Skip to main content

Full text of "Proceedings of the California Academy of Sciences, 4th series"

See other formats


PROCEEDINGS 

OF THE 



California Academy of Sciences 



FOURTH SERIES 



VOLUME XXXVIII 



ircstschrvjr 
for 



Jn J{mor ofJ(is 




SAN FRANCISCO 
PUBLISHED BY THE ACADEMY 

1970 



^ame Biological LEJcraiony 

_ LtBRARY ' 

FEB 1^< 1971 



:WOODS HOLE, f^ASS. 



-^ 




VVashinston Biologist's Field Club; 
Plummers Island, Md., 1933 

Photo by A. K. Fisher 



Retirement Photo 1970 




With HIH Crown Prince Akihito in the Stanford Collection 1967 

I'alo Alto Times Photo 



GEORGE SPRAGUE MYERS 






PROCEEDINGS 



OF THE 




California Academy of Sciences 



FOURTH SERIES 



VOLUME XXXVIII 



for 

Jn J(onor ofJiis/ixtu-Jifth ^irthdaii 



SAN FRANCISCO 

PUBLISHED BY THE ACADEMY 

1970 



COMMITTEE ON PUBLICATION 

George E. Lindsay, Chairman 
Edward L. Kessel, Editor Leo G. Hertlein 



CONTENTS 




Number 
1. 



3. 



7. 

8. 



10. 



11. 



12. 



Pages 
Introduction. By Earl S. Herald vii 

On the natural history of George Sprague Myers. By 

Lionel A. Walford 1-18 

Annotated chronological bibliography of the publications of 

George Sprague Myers (to the end of 1969) 19-52 

A new species of the doradid catfish genus Leptodoras, with 
comments on related forms. By James E. B5hlke 53-62 

Systematics of the genus Hemitriakis (Selachii: Carcharhini- 

dae), and related genera. By L. J. V. Compagno 63-98 

Notes on the natural history of snipe eels. By Giles W. 

Mead and Sylvia A. Earle 99-104 

The zoogeography of the herpetofauna of the Philippine 
Islands, a fringing archipelago. By Walter C. Brown and 
Angel C. Alcala 105-130 

Tropical shelf zoogeography. By John C. Briggs '. 131-138 

A new species of glandulocaudine characid fish, Hysteronotus 
my erst, from Peru. By Stanley H. Weitzman and Jamie 
E. Thomerson 139-156 

Rediscovery of the loricariid catfish, Acestridium discus 
Haseman, near Manaus, Brazil. By Robert L. Hassur 157-162 

The amphibians and reptiles of Afghanistan, a checklist and 
key to the herpetofauna. By Alan E. Leviton and Steven 
C. Anderson 1 63-206 

Explosive spread of the oriental goby Acanthogobius flavi- 
manus in the San Francisco Bay-Delta region of California. 
By Martin R. Brittan, John D. Hopkirk, Jerrold D. 
Conners, and Michael Martin 207-214 

Some nerve patterns and their systematic significance in 
paracanthopterygian, salmoniform, gobioid, and apogonid 
fishes. By Warren C. Freihofer 215-264 



CONTENTS 

13. A new gymnotoid fish from the Rio Tocantins, Brazil. By 
Maarten Korringa 265-272 

14. On the trail of the golden frog: with Warszewicz and Gabb 

in Central America. By Jay M. Savage 273-288 

15. Teleost hybridization studies. By Clark Hubbs 289-298 

16. A revision of the fishes of the genus Notothcnia from the 
New Zealand region, including Macquarie Island. By Hugh 

H. DeWitt 299-340 

17. How many recent fishes are there? By Daniel M. Cohen 341-346 

18. Description of a new subspecies of Rhabdophis aurkulata 
in the Philippines, with comments on the zoogeography of 
Mindanao Island. By Alan E. Leviton 347-362 

19. A reinterpretation of the teleostean fish order Gobiesoci- 

formes. By William A. Gosline 363-382 

20. Scale-eating American characoid fishes with special reference 

to Probolodus hctcrostomus. By Tyson R. Roberts 383-390 

2 1 . The dorsal and anal spine-locking apparatus of surgeon fishes 
(Acanthuridae). By James C. Tyler 391-410 

22. Amputation and replacement of marginal spines in ctenoid 

percoid scales. By Howard McCully 411-414 

23. Size and distribution of proteins in elasmobranch plasma. 

By Robert J. Heckly and Earl S. Herald 415-420 

Index 421-438 



INTRODUCTION 

The faculty appointment of George Sprague Myers as a Stanford associate 
professor in 1936 and full professor in 1938 marked the resurgence of what had 
informally been recognized as the Jordan school of ichthyology. In the thirty- 
four years since that time more than 104 graduate and special students as well 
as a large number of undergraduates, have come under Professor Myers' guid- 
ance. Although the majority of these students were involved in the study of 
fishes, a very respectable number specialized in amphibians and reptiles. During 
this same period the Stanford Ichthyological Bulletin came into prominence as 
did a series of important herpetological reports published as Occasional Papers 
of the Natural History Museum of Stanford University. 

On behalf of the many students and colleagues who have carried out their 
studies at the Stanford Natural History Museum, the authors participating in 
this volume respectfully dedicate this Festschrift to George Sprague Myers in 
appreciation of his helpful leadership in the field of systematics of the lower 
vertebrates. 

EARL S. HERALD 
August 1970 



PROCEEDINGS 

OF THE 

CALIFORNIA ACADEMY OF SCIENCES 

FOURTH SERIES 



Festschrift for George Sprague Myers 



Vol. XXXVIII, No. 1, pp. 1-18. December 31, 1970 



ON THE NATURAL HISTORY OF 
GEORGE SPRAGUE MYERS^ 

By 

Lionel A. Walford 

Sandy Hook Marine Laboratory 
Highlands, Neic Jersey 07732 

At a very young age, George Sprague INIyers manifested those qualities 
which were to remain his mark of distinction — an extraordinary sensitivity to the 
beauty of order in Nature, a boundless capacity to learn about what interested 
him, and a zest for collecting, arranging, and reasoning how things must fit to-^ 
gether. Given such an endowment, the place where he was born and spent his 
boyhood — Jersey City, New Jersey — and the epoch of his birth — early part of 
the twentieth century — were peculiarly right for guiding him towards and into 
his life work. For at that time many of the nineteenth century systematic 
zoologists were still flourishing (David Starr Jordan, for example) and there were 
plenty of roads from Jersey City leading to their doors and also to back country 
that was still unspoiled and beckoning. 

Jersey City in 1905, the year of Myers' birth, was already well established 
as part of what was to grow into the great Atlantic megalopolis. Like its neigh- 
boring satellite communities, it did not share any part of New York's splendor. 



1 1 am deeply grateful to Mrs. ^Mary S. McKenzie, for providing information on the family background 
and early history of her nephew; and also to Dr. Alan E. Leviton through whose good offices I have been 
permitted to consult and use several autobiographical fragments which Myers wrote at various times and 
deposited along with his extensive file of biographical information on zoologists at the California Academy 
of Sciences. 

[1] 



2 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

yet was close enough to attract the most unbeautiful features of industrial de- 
velopment. 

The house where Myers was born and spent much of his boyhood was a 
three-story brick structure at 283 Grove Street, directly opposite the front of the 
city hall. Even in 1905, this was part of a dwindUng genteel neighborhood in the 
process of being eaten away by slums that were surrounding it. 

The biota of the Grove Street house was typical of relic residential sections in 
northeastern crowded cities. Near the woodshed in the back yard grew a large 
Rose-of-Sharon bush, a lilac which annually put out a few flowers, some South 
American spider plants (Cleome) . There were a few Ailanthus trees in nearby 
back yards, and an ancient linden that grew out of a hole in the flagstone side- 
walk near the curb. Of insects that aroused some interest in young Myers, albeit 
a short-lived one, were bees of various sorts. "What could you do with a bee?" 
he asked. The only birds were English sparrows. Mammals consisted of cats, 
dogs, rats, and bats which issued at dusk from a nearby church. One of the cats 
was a pet, the first of a series which Myers has had almost continuously ever 
since. 

Myers' introduction to fish life was a minnow, probably Chrosomus neogaeus, 
living in a spring on a farm to which he was taken during a summer trip to 
Maine at the age of seven. Next he met some entrancing goldfish in a pet shop 
where his mother and Aunt Mary often took him during their visits to New 
York. He actually got to possess one or two goldfish in a bowl at various times. 
These experiences were no more remarkable than any other young city boy might 
have during his natural history phase. To young Myers, however, this was no 
phase, but rather a prologue to his great lifelong interest. This really began at 
age 12, when he first attended the Jersey City Aquarium Society's annual ex- 
hibition in the public library. Enthralled by the colorful fresh-water fishes 
from all over the world, he promptly joined the Society and began to accumulate 
aquariums in which he kept not only exotic fishes but also native ones ( Umbra, 
Funduhis, Enneacanthus, etc.). These he caught on trips made by train or 
trolley car to various places in northern New Jersey. At the same time he 
collected and kept amphibians and reptiles, these being as interesting to him as 
fishes. When he was about 15 years old, attending Lincoln High School in Jersey 
City, he sought advice of his biology teacher about a trip he was planning to 
the Pine Barrens of Lakehurst, New Jersey, to collect the beautiful rare tree frog 
Hyla andersonii. The teacher, rather out of his depth, suggested that Myers dis- 
cuss his problem with Dr. G. K. Noble at the American Museum of Natural His- 
tory in New York. By following that advice, Myers became introduced to the 
world of research zoologists. Noble, impressed, of course, introduced him to 
A. I. Ortenburger; and when the two had become well enough acquainted with 
this interesting young fellow, they took him on the last of Noble's Lakehurst trips 



Vol. XXXVIII] WALFORD: HISTORY OF GEORGE S. MYERS 3 

to Study the life history of Hyla andcrsonii. Through Noble, Myers came to 
know Karl P. Schmidt, and, in the Museum's fish department, John Treadwell 
Nichols, Arthur W. Henn, and Eugene W. Gudger. It was not long before he 
was spending so much time on his cold-blooded vertebrates that his school work 
slipped badly, for he took full advantage of the proximity of New York to meet 
most of the old guard zoologists at the Museum, the Zoo, and the Aquarium — 
Henry Fairfield Osborn, Bashford Dean, F. A. Lucas, Walter Granger, W. D. 
Matthew, Frank M. Chapman, Carl Akeley, Robert Cushman Murphy, Roy 
Chapman Andrews, C. H. Townsend, William Beebe, John Tee-Van, and Charles 
M. Breder. i\nd when a young fellow from the University of Virginia was 
selected by Noble to go with Andrews to collect reptiles in China, Noble and 
Myers went out to Plainfield, New Jersey with him to teach him how to collect 
salamanders. His name was CHfford H. Pope. 

One day in 1924 while at the Museum, Myers was introduced to Dean Carl 
H. Eigenmann of Indiana University, the principal worker on the systematics of 
the fresh-water fishes of South America. Myers had then published a few short 
papers on fishes and had become especially interested in those of tropical Amer- 
ica. The result was an invitation by Eigenmann to come to Indiana as a student 
and have part of the cost defrayed by part-time work in caring for the Indiana 
fish collection. 

IVIyers had not done well in high school, and lacked several credits for grad- 
uation. Moreover, with the examples before him of several then well-known zool- 
ogists who had had no university preparation, he was uncertain even whether to 
go to college. However, he says that Noble gave him a thorough tongue-lashing 
about his refusal to get the necessary schooling, and this, coupled with Eigen- 
mann's offer, decided him to go. In lieu of his missing high school credits. Dean 
Eigenmann arranged to have him granted credit for the research he had already^ 
accomplished. He has always felt a great deal of gratitude to Noble and Eigen- 
mann, without whose help his professional career might have been aborted. 

At Indiana, he neglected his academic studies to accomplish some field work 
and to complete a synopsis of the amphibians and reptiles of Indiana. At the 
same time he got a superb introduction to South American fishes, and also to 
curatorial methods for preserved research collections. Aside from Eigenmann, 
the man on the Indiana faculty who had the most influence on Myers (though 
Myers took no formal courses from him) was the entomologist. Professor Alfred 
C. Kinsey (later the student of human sexuality), whose forward-looking views 
of evolution and systematics were then finding expression in studies of cynipid 
wasps. From Kinsey, Myers began to gain a very broad view of systematics as 
a synthesis of the comparative aspects of all other biological disciplines, a view 
that finally found expression in a review paper in 1930. 



4 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

After Myers had been at Indiana one year, Eigenmann fell ill and was taken 
first to Florida and then to the San Diego region of California for his health. 
Since he obviously would not be able to return to the University, there seemed 
little reason for Myers to remain. However, he unpacked the large Ternetz 
South American collections as they arrived, and, with Eigenmann's permission, 
described some of the novelties. 

Stanford University, founded in 1891, had become a center of research on 
fishes and their habits through the influence of the university's first president, 
David Starr Jordan, and the first chairman of its zoology department, Charles 
H. Gilbert. In 1926, both men had been long retired, but Jordan was still active 
and he had noticed the papers published by Myers. When he heard of Eigen- 
mann's illness, Jordan wrote to Myers asking whether he would like to transfer 
to Stanford with the same sort of part-time assistantship he had had at Indiana. 
Jumping at the opportunity, Myers arrived in California in September, 1926. 

It was shortly after his arrival at Stanford that I first met him there. As a 
biologist at the California State Fisheries Laboratory at Terminal Island, I had 
just begun a study of the California barracuda, and took advantage of an op- 
portunity to spend a few days at the Natural History Museum to search its 
library and to study the sphyraenid material in the fish collection. The Museum 
at that time was a center of quiet excitement such as I will never forget, for many 
of the biology faculty members focused their interests there, and the principal 
interest was the study of fishes. There is a vivid picture in my mind's eye of a 
pleasant, fine-looking, young man, an undergraduate, working as an assistant in 
the Museum. "May I help you?" he asked. "Perhaps," I replied, cataloging him 
as a library assistant. When I told him what I wanted, he poured a steady 
stream of information out oj his head: "Of course, you must already have the 
essential papers about Sphyracna argentea and S. ensis. You may have missed 
Sphyracna idiastes. We have the original 1903 description by Heller and Snod- 
grass." And so on through the whole family Sphyraenidae around the world. 
No, he was not working on sphyraenids himself. Never had. It seemed to me 
as I talked to this enthusiastic modest fellow that he knew everything about 
everything. He was already a learned person when most people are scratching 
about, trying to make up their minds what they want to learn about. 

The area surrounding the Stanford lands was then open country, and the 
University community (affectionately known as "The Farm" by faculty and 
student alike) moved at a relatively leisurely pace. Under Jordan's influence it 
had been, and still was, a great center for studying systematic zoology, especially 
fishes. Jordan was still working on fossil fishes. John Otterbein Snyder was 
chairman of the zoology department, with Edwin C. Starks as morphologist, 
Harold Heath as invertebrate zoologist, G. C. Price as embryologist, and Isabel 
McCracken, R. W. Doane, and G. F. Ferris as entomologists. Aside from courses 



Vol. XXXVIII] WALFORD: HISTORY OF GEORGE S. MYERS 5 

in other departments, Myers came to know Snyder, Starks, Heath, and Ferris 
best, as well as H. G. Schenk and S. W. Muller in the geology department, in 
which he eventually took his doctorate minor. 

The Dudley Herbarium, the Entomological Collections, and the Zoological 
Collections were then housed — temporarily, it was said — in the south end of the 
Stanford Museum where they remained as the "Natural History Museum," and 
later the "Division of Systematic Biology," for the entire period of Myers' active 
association with the University. Professor LeRoy Abrams was in charge of the 
Herbarium, Ferris of the insects, and Snyder of the Zoological Collections. 

Although Snyder was his special mentor and friend, Myers visited Jordan, 
at Jordan's warm invitation, at least once a week. Snyder, like Gilbert before 
him, had almost deserted systematics to work on the migrations of salmon and 
steelhead trout. In 1928, Willis H. Rich was appointed to the department, to 
teach ecology and fishery biology. 

A biweekly seminar in fishery biology in 1928-29, attended by Snyder, Rich, 
Frank Weymouth, Starks, and about a dozen serious students including Myers, 
most of whom were to become leaders in the study of fishes, was a whirlwind of 
lively discussion and argument among and between students and professors, such 
as I have rarely experienced since. 

Because the Stanford group then most interested in the theory of systematics 
was led by Schenk in geology, Myers gravitated to that quarter. In connection 
with one of Schenk's seminars, Myers published a review of a recent botanical 
revision in Schenk's Micropaleontology Bulletin, in which the ideas he had de- 
veloped after contact with Kinsey at Indiana and Schenk at Stanford were 
synthesized into a view of systematic biology that was unusually broad for its 
day. When it is recalled that Myers was then an undergraduate student, without 
much knowledge of what was then being done on theoretical systematics by 
several isolated men or groups in America and Europe, his statement is remark- 
able. 

Myers' university work went slowly, not only because of his part-time em- 
ployment but also because he was able only slowly to force himself to neglect 
extracurricular work in ichthyology and herpetology enough to get good marks. 
One objection to granting him a bachelor's degree was that he had not taken 
enough required courses in English. Snyder later confided to me (with much 
amusement), that he had demolished this hurdle simply by waving in the faces of 
the objecting English Department's faculty members a handful of the publica- 
tions Myers had produced since he entered college, saying, "Look, see how much 
he has written!" He had published a good deal by the time he was granted his 
A.B. degree in June, 1930, seven years after he entered Indiana as a freshman. 
After that, things went more rapidly. He obtained an A.M. in 1931, and his 
Ph.D. in June, 1933. 



6 CALIFORNIA ACADEMY OF SCIENCES LProc. 4th Ser. 

Of his Stanford student years, Myers has said that he learned more mor- 
phological zoology from Harold Heath and Tage Skogsberg at the University's 
Hopkins Marine Station (where he spent the summer of 1929), than from all 
others. Nevertheless, Snyder, Starks, and Rich ranked high in his training years. 
For four years he saw and talked with Jordan almost weekly, gaining a great 
wealth of information about ichthyological workers and history. 

In 1928, Ur. Albert W. Herre, a former pupil of Jordan, was appointed to a 
non-faculty position as Curator of the Zoological Museum at Stanford with an 
arrangement by which he would be retired only upon the President's pleasure. 
Myers says that he owes much broadening and maturing to Herre's influence, 
not only while he was a student, but also after he returned to Stanford in 1936. 

In 1933, Myers was appointed Assistant Curator in charge of the Division 
of Fishes at the U.S. National Museum, with a first assignment to pack and ship 
many National Museum fishes that had been at Stanford in the hands of the late 
Charles H. Gilbert. He arrived in Washington in March, 1933. 

At the National Museum, Myers took charge of the most important ichthy- 
ological resarch collection in America but one which had suffered from nearly 40 
years of impoverishment and neglect. Moreover, the great financial depression 
of the 1930's had worsened the situation so much that for the first two years of 
Myers' tenure, the Division lacked even the services of a typist. Besides Myers, 
the staff consisted only of one elderly but enthusiastic scientific aid. Earl D. 
Reid-, and a laborer who cared for the alcoholic collections. For the next three 
years, Myers and Reid spent most of their time in sorting, bottling, and register- 
ing an enormous backlog of specimens and putting the Division's offices, files, 
and records into working condition. During most of this period they trained and 
supervised squads of up to a dozen temporary workers at a time, these having 
been supplied free to government bureaus by successive federal agencies set up 
to relieve unemployment. 

Although there was precious little time for research during these three years, 
Myers and Reid initiated a survey of the fresh-water fishes of Virginia. The only 
help from the impoverished Museum was for bottles and alcohol, but using Reid's 
old automobile and paying all other expenses themselves, they made many collec- 
tions from the Dismal Swamp to the mountains of western Virginia. 

Myers says that in Washington he had the best and most cooperative superior 
administrators that a curator could have. His immediate superior was the late 
Leonhard Stejneger, Head Curator of Zoology and Curator of the Division of 
Reptiles and Amphibians, whose kindliness and enormous memory he has always 
remembered with pleasure. The other was Alexander Wetmore, then Assistant 



- Myers always expressed the greatest admiration for Reid, who was invalided out of the U.S. Marine Corps 
after being wounded in the eyes while in Nicaragua. Reid became a doorman in the Museum and worked his 
way to the Civil Service subprofessional grade of Aid by taking night courses in zoology at George Wash- 
ington University. 



Vol. XXXVIII] WALFORD: HISTORY OF GEORGE S. MYERS 1 

Secretary of the Smithsonian and Director of the National Museum, whose 
administrative abihty and thouf^htfuhiess for his staff were boundless. 

In 1936, Myers was invited back to Stanford, and after considerable thought, 
accepted a position as Associate Professor of Biology and Head Curator of 
Zoological Collections, with the provisions that he be advanced to Professor by 
1938, that half his time be spent on curatorial duties and half on teaching, and 
that Dr. Herre's employment as Curator of Zoology not be terminated as the 
department head had planned. With Herre's retention assured, Myers assumed 
his new position in September, 1936. 

George Myers has told me many times that his most important contributions 
to ichthyology and herpetology have been the help and guidance he has been able 
to give to the long line of graduate students who worked with him at Stanford. 
When he began teaching in 1936, no formal course dealing with more than the 
barest rudiments of taxonomic ichthyology appears to have been given anywhere. 
In those days, prospective taxonomists were supposed to pick up knowledge of 
their field without any formal guidance. The Stanford fish course initiated by 
Charles H. Gilbert and continued by John O. Snyder had consisted solely of 
identifying specimens with the aid of Jordan and Evermann's "Fishes of North 
and Middle America." Myers has told me that his own background in ichthy- 
ology and vertebrate evolution was very defective as a consequence, so that be- 
tween 1936 and 1938 he found it necessary to prepare himself by doing a great 
deal of reading and studying. The books of Goodrich and Romer and the papers 
of C. Tate Regan proved to be of the greatest help. In 1938, for the benefit of a 
small group of students, including W. A. Gosline and E. S. Herald, he attempted 
a general summary of fish classification and evolution, with emphasis on the 
literature and history and on major groups down to the family level. This first 
attempt developed into a more formal course, called "Advanced Systematic 
Ichthyology," which was usually given every other year, alternating with a 
shorter, somewhat less advanced course in herpetology. This course in ichthy- 
ology formed the genesis of other more or less similar courses, such as that given 
by W. A. Gosline first at the University of Michigan and later at the University 
of Hawaii. Myers also gave annually a course at first called "Vertebrate 
Paleontology" and, later, "Evolution of the Vertebrates," and a short course on 
zoogeography. Myers' most popular course, planned and given by himself and 
his botanical colleague. Professor Ira L. Wiggins, was a general survey of plant 
and animal ecology, including ecology of man. It was given for two or three 
years in the late 1940's primarily for non-biologists. In this course, Myers was 
one of the first to emphasize the rapidly increasing danger to the human race 
caused by the unrestricted growth of human population. Unfortunately, other 
work forced Wiggins to withdraw from the course, and, as Myers felt himself 
incompetent to handle the botanical side, the course was regretfully dropped. 



8 C ALIFORM A ACADEMY OF SCIENCES [Proc. 4th Ser. 

Myers has maintained that graduate teaching is greatly helped by the 
presence of a "critical mass" of at least four enthusiastic students working under 
one professor on different dissertations in related fields. The presence of such 
a group was barely attained in his laboratory when World War II intervened. 
Students vanished, and he was sent to Brazil for two and one-half years. It was 
not reattained until the late 1940's; but from that time until well into the 1960's 
a "critical mass" was continuously present, rising at one point to as many as 12 
or 13, usually two or three of them herpetologists and the rest ichthyologists. 
As he says: "Those were the most exciting and rewarding years that I have 
experienced. As a matter of principle, no graduate student was assigned a doctor- 
ate problem, or encouraged to choose one closely related to any of my own re- 
search. They were forced to select their own, my only requirements being that 
the problem be reasonably interesting and difficult but not impossibly time- 
consuming, and that it be concerned with areas within which I felt myself fairly 
competent to judge quality of performance. The atmosphere was never dull. 
Everybody helped and taught everybody else, the professor learning as much 
from the students as they did from him. Chores in the old Natural History 
JNIuseum, such as registering and care of material, helping with the editing of the 
Stanford Ichthyologiral Bulletin, and the like, were often done almost as much 
by those who were not paid to do the work as those who were. My own contribu- 
tion was largely that of arbiter, critic, walking bibliographer, ruthless editor of 
often poorly expressed, first attempts to write up scientific results, father-con- 
fessor, cheerleader, and especially as the provider and keeper of a laboratory 
atmosphere conducive to hard work, cooperation, enthusiasm, and high attain- 
ment. There was little of the formality that often separates professor from stu- 
dent. Evening seminars or meetings often ended in a nearby Bierstube, and I 
was usually invited to student parties. Former students have often remarked on 
the uniqueness of human relations in the Museum and recall them with nostalgia 
— as do I. Yet, I must have commanded a modicum of respect for I have noted 
with some amusement that none of my former graduate students ever tried to 
address me by my first name (a common enough thing in the U.S.A.) until 10 or 
IS years after obtaining their doctorates: and several have never been able to 
bring themselves to do so. But when the number of graduate students rose to 
eight, ten, or a higher number, I got comparatively little research of my own 
done, for I was available to all of them almost every day — an arrangement at 
which some other groups of graduate students, both at Stanford and elsewhere, 
marvelled. Yet near the height of my graduate student load, economic necessity 
forced me to write extensively for — and manage — a popular aquarium magazine 
{The Aquarium Journal). This very difficult regime went on for two years 
(1952-54) until my Stanford salary rose enough to make it possible to give up 
most such writing." 



Vol. XXXVIII] WALFORD: HISTORY OF GEORGE S. MYERS 9 

The Stanford fish collection was ori,ginall3' small, consisting largely of dupli- 
cates from the field work of Jordan and his pupils. Through the years it grew 
slowly through an unexpressed policy of growth in diversity without amassing 
long series of individual species, a policy which Myers enforced in more recent 
times. It gradually attained a diversity among American collections second only 
to the National Museum, although in numbers of specimens (between 750,000 
and 1,000,000) smaller than several other large research collections. It has been 
especially useful for morphological work in systematics such as has been em- 
phasized in recent years. 

The herpetological collections were small when Myers took charge in 1936, 
consisting of fewer than 2,000 amphibians and 10,000 reptiles, mostly collected 
during early work in the days when John \'an Denburgh was a student. Myers 
built these collections up judiciously until now they total about 60,000 speci- 
mens, half amphibians and half reptiles — numerous enough and diverse enough 
for many systematic purposes. 

For curatorial work and the management of the Zoological Collections pri- 
marily as a laboratory for graduate students, Myers was greatly aided by 
Margaret H. Storey. She had obtained her A.]M. degree with Willis Rich while 
Myers was at the National Museum, had stayed on as a volunteer assistant later 
to be appointed Assistant Curator of Zoological Collections. She supervised the 
paper-work and curating, helped edit Stanjord Ichthyologkal Bulletin and was 
a tower of strength and help to all those who worked in the Zoological Museum 
until her untimely death in 1960. 

Myers and Storey together worked out systematic methods, some of them 
new, for sorting, registering, bottling, labelling, arranging, installing, and finding 
bottled museum specimens. These methods, described chiefly in three of the 
Museum's mimeographed circulars, made it possible for much of the work to be 
done by untrained student helpers, and to handle a large research-collection 
operation (up to a million specimens) with less than half the staff and funds 
usually available for such purposes. 

For all of ISIyers' years on the faculty, the Zoological Collections had no 
more than four employees besides himself — A. W. Herre, until World War II, 
Margaret Storey and later Warren Freihofer, as aid or associate curator, one 
half-time student assistant, and, after World War II, one typist-secretary who 
also served entomology. This staff handled large and growing research collec- 
tions of fishes, amphibians, and reptiles as well as sizable collections of 
mammals, birds, and aquatic invertebrates. They were also responsible for 
the time-consuming processing of extensive loans to researchers elsewhere, 
running from about 500 to as many as 5,000 specimens annually. In one thing, 
Myers was adamant. Collections of animal groups in which he had no direct 
interest were also kept in good condition and order. Such curatorial conscience 



10 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

is rare. Moreover, collections which had strayed away from the Museum to 
other parts of the University and elsewhere were retrieved and set in order. 
Types were rigidly labelled and segregated with blue (holotypes) or red labels. 
Species presumed to be extinct received green labels. 

The library on Systematic Ichthyology at the Museum was rich in the older 
literature and in reprints, all from Jordan's personal library. It was kept up by 
exchanges for Stanford Ichthyological Bulletin and by judicious buying with the 
small funds available. Concurrently, Myers personally purchased many ichthy- 
ological and herpetological books that were not present in the Museum, and his 
library books and reprints admirably supplemented the Museum's holdings. To- 
day, since Professor Carl L. Hubbs' library has gone to Scripps Institution, 
Myers probably has the most extensive private fish library in the world. It is 
especially rich in reprints but lacks such expensive items as the great works of 
Bloch, Bleeker, and Agassiz. 

Myers started Stanford Ichthyological Bulletin in 1938. It was printed 
cheaply by offset (the text being typed by the Museum staff), since funds for 
this journal were always miniscule. They began at $133.00 annually and never 
rose over $750.00. All sorts of schemes to get outside funds were tried, usually 
with only moderate success. Of the eight volumes that eventually appeared, less 
than half were or could be paid for from regular funds. 

Myers and Miss Storey were the principal movers in two local groups. The 
old "Stanford Zoology Club," which originated in the 1890's and was supported 
by generations of Stanford students, was revived as the "Natural History Club" 
and survived until the 1950's. A new, informal group, the "Fishverein," com- 
posed of those at Stanford interested in fishes and the many local fishery biolo- 
gists working for the Federal Fish and Wildlife Service and the California 
Division of Fish and Game, was formed by Myers and met fairly regularly for 
many years. 

During his early preuniversity years (1920 24), Myers' papers reflect the 
growing interests and ability of an untrained young man deeply interested in 
the habits and taxonomy of the lower vertebrates. He published his first articles 
on aquarium fishes at the age of 15, in 1920. These early attempts give an 
inkling of the extensive boyhood observations representing dozens of families of 
live fishes, and also amphibians and reptiles, either in captivity or in the field in 
New Jersey and North Carolina. As Myers says: "By the time I was 19, I knew 
in a general and sometimes specific way a great deal about fish behavior that has 
of late been 'discovered' and formally categorized by the fish behaviorists, in the 
same way that the field ornithologist becomes familiar with bird behavior." 

By the end of 1923, Myers had published his first really scientific papers, 
one on a new poeciliid from Hispaniola with J. T. Nichols and others on the 
nomenclature of anabantids. By the end of 1924, he had published nine tax- 



Vol. XXXVIII] WALFORD: HISTORY OF GEORGE S. MYERS 11 

onomic papers on fishes, and one herpetological paper. It was in 1924 that he 
made his first longer field trip, to Wilmington, North Carolina, where he made 
many observations and discovered what is now known as the common dusky 
shiner {Notropis cummingsi Myers) of the southeastern coastal plain, the de- 
scription of which he published in 1925. 

Although the beginning of university work in 1924 curtailed his output of 
papers, he continued publishing on a variety of ichthyological and herpetological 
subjects up to the time he finished his schooling at Stanford and went to the 
Smithsonian in 1933. To refer to only a few of the papers which he published 
during his student years at Indiana and Stanford (1924-33), there is a synopsis 
of Indiana amphibians and reptiles (1926), four papers on amphibians (1930- 
31), descriptions of many South American fishes collected by Ternetz (1927), 
a revision of the genera of neotropical cyprinodontids (1927). three or four im- 
portant papers on Chinese fishes, and a prophetic paper on the phallostethids 
which foreshadowed some of the important features of Rosen's radical reclassifi- 
cation of the atheriniform fishes in 1964. In addition, Myers found time in 1929 
to write a sizable addendum to the final volume of Eigenmann's 'The American 
Characidae." 

IMyers was faced with such exceptionally time-consuming curatorial duties 
at the Smithsonian that his research during those years (1933-36) suffered. 
However, he reviewed the genera of triacanthids in 1934, published on the cy- 
prinodonts of Hispaniola as well as the opistognathids (and owstoniids) in 1935, 
and revised the genera of Polynemidae in 1936. In that same year, in a report 
on fishes from Lake Tanganyika, he briefly pointed out for the first time some 
of the strange features of lake-fish evolution. 

After beginning his teaching and curatorial work at Stanford in 1936, Myers' 
first paper was one that he had read before a meeting in New York in 1934 and 
which he based on observations made in the 1920's. In this short paper, he 
arrived independently at the same conclusions as had C. M. Breder, Jr. in regard 
to the evolution of oral brooding in cichlid fishes. 

The most widely known and influential of INIyers' papers, prepared for the 
1937 Smithsonian Report (1938) was his "Fresh-water Fishes and West Indian 
Zoogeography." He had been highly dissatisfied with most writings on historical 
zoogeography, particularly the prevalence of the ideas of Matthew and others 
based largely on the tetrapod evidence, and especially with the use made of the 
evidence of fresh-water fishes. In this paper, dealing specifically with the West 
Indies but ranging over the fishes of all continents, Myers pointed out that what 
had previously been taken for true fresh-water fishes are really divisible into two 
physiologically different groups, one with considerable salt tolerance and the 
other ("primary fresh-water fishes") much more strictly confined to fresh water. 
The primary fresh-water fishes are much less able to spread across continents 



12 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

and sea gaps than are mammals and even amphibians, and thus their dispersal 
patterns provide a much more conservative and dependable guide to the past 
existence of these gaps than do those of tetrapods. Myers' zoogeographical con- 
clusions, although stated only cautiously and tentatively, agreed with those of 
Matthew in regard to the absence of past continental connections of the West 
Indies, but disagreed with Matthew in the strong evidence provided by the 
primary fresh-water fishes for a past southern trans-Atlantic connection. It is 
notable that 20 years after publication this paper was acknowledged by P. J. 
Darlington in his great book "Zoogeography," as the prime reference on which 
he built that part of his book dealing with fishes. Myers' 1938 work combined 
with his later papers on salt tolerance of fresh-water fishes (1949) and East 
Indian zoogeography (1951), gave new direction to later studies on the his- 
torical zoogeography of continental vertebrates. Myers seemed more than half 
convinced of the truth of continental drift in 1938, and although he faltered in 
that conviction in his 1951 paper, he later reaffirmed it in 1966 and 1967, be- 
cause by then the weight of his evidence favored the primary fresh-water fishes 
as the most significant vertebrate indicators for establishing past continental 
connections. 

Early in 1938, Myers was able to accompany that year's expedition of the 
Allan Hancock Foundation's ship Velero III to the coasts of Mexico, Cocos Is- 
land, the Galapagos, Peru, Ecuador, and Panama. This resulted in collaborative 
papers with C. B. Wade on eels (1941), atherinids (1942), and other fishes 
(1946). In addition, a study on the zoogeography of the fishes of the Pacific 
Ocean appeared in 1941. 

Herpetological work had been impossible in the Division of Fishes in Wash- 
ington; but on Myers' return to Stanford he began a number of smaller studies 
on amphibians and lizards which culminated in six herpetological papers in 1942. 
One of these described the now well-known black toad of Deep Springs Valley 
{Bujo exsid Myers), which has perhaps the smallest range of any living am- 
phibian. 

Following the entrance of the United States into World War II, Myers was 
posted to the Museu Nacional in Rio de Janeiro, as part of a governmental plan 
to maintain good relations with Latin America in troubled times. He arrived in 
July, 1942, for a one-year period, which eventually lengthened to nearly 2% 
years. In Rio he helped with curatorial and library methods, with setting up civil 
service categories for the museum staff, with exhibits and with museum admin- 
istration. For the federal fish and game division and the Sao Paulo fish and 
game department, he helped by devising better methods of gathering fish-catch 
statistics. In addition, for a period of over a year, the Museum lent his services 
to the Rio office of the U.S. Coordinator of Inter-American Affairs. There was 
little time for research, and the wartime shortage of gasoline made travel by 



Vol. XXXVIII] WALFORD: HISTORY OF GEORGE S. MYERS 13 

automobile next to impossible. Nevertheless, he managed to take many local 
trips, principally by tramway on weekends, to the wilder areas in the metro- 
politan region. These trips were mostly for frogs, in the company of Dr. Bertha 
Lutz and Joaquim Venancio or Antenor Carvalho. Eventually, there were longer 
trips with Carvalho or others by train and other conveyance, to the Rio Sao 
Francisco at Pirapora, to Santa Teresa in Espirito Santo, and southward along 
the coast to Rio Grande do Sul. Papers resulting from the Brazilian years were 
few, most of them appearing in 1944 and 1945. 

On Myers' return in 1944, he hoped that the survey of Brazilian marine 
market fishes that he had helped to originate would result in taxonomic studies 
of these fishes at Stanford by Brazilian students; but the students did not ap- 
pear and the project languished after 1950. The sole results have been the 
amassing of an excellent representation of Brazilian shore fishes in the INIuseu 
Nacional, and a smaller duplicate set at Stanford. 

A trip to attend the Pacific Science Congress in New Zealand in 1949 re- 
sulted in two zoogeographical papers, one on East Indian fishes (1951 and 1954, 
published twice) and the other on East Indian amphibians (1954) both of which 
tended to firm up the concept of Wallace's Line. Myers had become editor of 
an aquarium magazine for two years in the early 1950's. Several of the articles 
published then have ichthyological interest, chief of them being "Annual Fishes" 
(1952), which brought together and greatly strengthened by original observa- 
tions what had consisted of scattered and mostly nonscientific reports of tropical 
cyprinodontid fishes which exhibit a diapause when no individuals are alive ex- 
cept as zygotes. 

At the 1958 International Congress of Zoology in London, Myers presented 
a paper on the endemic fishes of Lake Lanao having an important bearing on 
evolution. In this paper, published in 1960, he was able to show that this 
cyprinid fauna, now diverse enough to be alloted to several genera, almost cer- 
tainly evolved very rapidly from a single ancestral species, perhaps within 10,000 
years. He also pointed out similarities in the evolution of other lake faunas, and 
was able to establish an evolutionary sequence: 1) an increasing number of very 
similar species belonging to a single genus, culminating 2) in a ''species swarm;" 
then 3) the differentiation of a few species into new endemic genera, and finally 
4) considerable reduction in the total number of species. Thus the number of 
species of the large genus gradually diminishes while the number of distinctive 
endemic genera increases. Myers also pointed out the strong possibility that on 
a grand scale the evolution of Amazonian fishes and of deep-sea fishes might 
parallel that of lake fish faunas and indeed, the original evolution of the animal 
phyla. 

In the 1960's, Myers returned to zoogeographical studies of fresh-water fishes. 
His paper on the North American fauna (1963) was published only in an ab- 



14 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

stract which lacked the section on continental drift; but his 1966 paper on the 
derivation of the fresh-water fishes of Central America directly opposed Darling- 
ton's idea that the ancestors of the South American fish groups originated in 
Holarctica, and suggested continental drift as an answer. In 1967 appeared his 
"Zoogeographical Evidence of the Age of the South Atlantic Ocean," a brief 
exposition of his belief that the cypriniform fishes had originated in a South 
Atlantic continent which split in the Triassic or Jurassic to form the South 
Atlantic Ocean. In 1966 was published a collaborative work by Greenwood, 
Rosen, Weitzman, and Myers, "Phyletic Studies of Teleostean Fishes, with a Pro- 
visional Classification of Living Forms," which broke strongly with traditional 
classifications of the teleosts. 

Myers intended his series of apparently not directly related studies on fish 
zoogeography (1938, 1949, 1951, 1963, 1966, 1967), together with his two 1960 
papers on lake fish evolution and the 1966 collaborative teleost study, to form 
an integrated whole indicating as nearly as can be done at present how and when 
the ostariophysan (and particularly the cypriniform) fishes evolved and dis- 
persed. In these papers the problem is attacked from several directions on the 
basis of the living world fauna and the few known fossils, ecological constitution 
of the fishes, their probable place and time of origin from the salmoniform 
fishes, their dispersal and evolutionary patterns as seen against the background 
of paleogeography, all within the strictures imposed by the greater known fossil 
evidence derived from tetrapods. Considered in this way, the nine papers con- 
cerned form an impressive contribution to knowledge of the fresh-water fishes 
of the world. 

One thing that Myers has said of his papers is that not many of them are as 
important, or represent as much thinking, as do a number of his reviews, mostly 
published since 1930 in Copeia. Many taxonomic and other conclusions first 
appeared in these reviews. Moreover, the column called by Myers "Phylax 
Telescopus," which he maintained for a couple of years in Copeia during the 
1960's, contains some of the best biological criticism that has appeared any- 
where. Myers has said to me that if he is remembered for anything, he would 
like it to be for just a few things — his graduate pupils, his critical comments and 
reviews, his early espousal of the need for curtailment of human population 
growth, his pioneer urging of the conservation of non-food and game fishes, and 
his integrated series of papers on the evolution and dispersal of fresh-water fishes. 

Despite the number of publications listed in his bibliography (nearly 600), 
I doubt that he ever engaged in any research simply to increase the quantity of 
his publications. He has always avoided humdrum taxonomic questions unless 
they were of some special significance, for he is completely devoted to seeking and 
elucidating principles. Thanks in large measure to his scholarly creativeness, as 
well as to his subtle and boundless patient teaching, systematic ichthyology is 



Vol. XXXVIIIl WALFORD: HISTORY OF GEORGE S. MYERS 15 

alive and well today and the subject of vigorous teaching in many centers of 
learning where it is appreciated. It is a pity that Stanford has not appreciated 
the tradition it had inherited through Jordan or the treasure which he started 
in the Museum collection and libraries, and which INIyers built up and organized. 
Instead, the university authorities have callously determined to give this treasure 
away and discontinue — discontinue — further teaching in this field! This is par- 
ticularly tragic at a time when the natural history of the earth and its resources 
is the most important thing we can know. 

CHRONOLOGY 

1905 Born February 2, Jersey City, New Jersey, son of Harvey Derwood Myers and 

Lily Vale (Sprague) Myers. 
1911-18 Public elementary schools, Jersey City. 
1918-24 Public high schools, Jersey City. 
1919-20 St. John's Military School, Ossining, New York. 
1922-24 Association with American Museum of Natural History, especially G. K. Noble 

and J. T. Nichols. 
1924 Field work during May in vicinity of Wilmington, North Carolina. 

1924-26 Indiana University, with Carl H. Eigenmann. Curatorial assistant, fish collection. 
1926 Married Martha Ruth Frisinger, Decatur, Indiana, September 25. 

1926 Entered Stanford University, October. Beginning of association with D. S. Jordan, 

C. H. Gilbert, J. O. Snyder, E. C. Starks. 
1926-31 Museum assistant, Stanford. 

1929 Field work during April-June in western Texas and Arizona with Gregory M. 

Kranzthor. Rediscovery of Elaphe bairdii. 

1930 Field work in Death Valley — Amargosa region — with Joseph H. Wales. Discovery 

of Cypyinodon diaboUs. 

1930 Bachelor of Arts, Stanford, June. 

1931 Master of Arts, Stanford, June. 

1931-32 Teaching assistant in comparative anatomy, Stanford. 

1932-60 Associate editor. The Aquarium, Philadelphia, edited and published by William- 

Thornton Innes, also scientific editor, 19 successive editions of Innes' "Exotic 

Aquarium Fishes." 
1933 Appointed Assistant Curator, in charge. Division of Fishes, U.S. National Museum, 

Smithsonian Institution, Washington, D.C., January 1. 
1933 Doctor of Philosophy, Stanford, June. 

1934-36 Field work, freshwater fishes of Virginia, with E. D. Reid. 

1935 Birth of first child, Thomas Sprague Myers, Washington, D.C., August 28. 

1936 Awarded Silver Medal of the "Societe National d'Acclimatation," Paris, for work 

on acclimatization, habits, and taxonomy of exotic aquarium fishes. 

1936 Resignation from Smithsonian. Appointed to faculty. Department of Biological 

Sciences, Stanford University, as Associate Professor and Curator of Zoological 
Collections, September. 

1937 Birth of second child, John William Myers, Palo Alto, California, December 15. 

1938 Member, Hancock Pacific Expedition, aboard M. V. Velero III, from January- 

March, visiting coasts of Mexico, Guatemala, Cocos Island, Galapagos Islands, 
Ecuador, Peru, Chinchas Island, Gorgona Island, Colombia, Panama. 



16 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

1938 Co-leader, with Rolf L. Bolin, of Crocker-Stanford Deep-sea Expedition, aboard 

yacht Zaca, off California coast in September. 
1938 Initiated Stanford Ichthyological Bulletin. Editor to end of volume 8 in 1967. 

1938 Advanced to full Professor, Stanford, September. 

1939 Member, Fishery Organizing Committee, 6th Pacific Science Congress, Berkeley, 

California. 

1940-41 Intensive extracurricular work with William Allen White's "Committee to Defend 
America by Aiding the Allies." 

1942 Elected Corresponding Member, Zoological Society of London. 

1942-44 Posted to Rio de Janeiro (State Department funds) to aid Museu Nacional and 
Divisao de Caca e Pesca. Lecture course on ichthyology and fishery biology in 
Rio. Brief visits en route to Mexico City, Guatemala, Panama, Call, Bogota, 
Mariquita, Lima, Arequipa, Santa Cruz (Bolivia), Corumba. Intermittent field 
work near Rio, and (with Antenor Carvalho and others) to Minas Gerais, 
Espirito Santo, Sao Paulo, Parana, Santa Catarina, Rio Grande do Sul, and 
Belem do Para. 

1944 Return to Stanford, October. 

1945-51 Vice-President and Council Member, California Academy of Sciences, San Francisco. 

1946 Beginning of post-war upswing in graduate-student enrollment at Stanford. 

1947 Bikini Scientific Resurvey, U.S. Navy, aboard U.S.S. Chilton. Field work on 

Bikini and Rongerik atolls. Plankton Survey, Bikini lagoon. Visits to Kwajalein 
and Honolulu, summer. 

1949 Pacific Science Congress, Aukland and Christchurch, New Zealand. Some fish and 

reptile collecting on South Island and Aukland Harbor. Visits to Hawaii, Samoa, 
Noumea, Canton Island and Johnston Island en route. 
1949-51 President, American Society of Ichthyologists and Herpetologists. 

1950 Brief trip to Brazil during August and September, visiting Recife, Salvador (Bahia), 

Rio, Belem do Para, Manaus, and Puerto Rico. 
1951-53 Special taxonomic work, U.S. Fish and Wildlife Service, Washington, D.C., summers. 
1952-54 Managing editor. Aquarium Journal, San Francisco. 
1954 European trip for Fish and Wildlife Service and FAO. Paris, with stop in London, 

December. 
1958 Field work and fish collecting during February, upper Rio Caqueta basin, vicinity 

of Tres Esquinas, Colombia, with General Thomas D. White. Visit to Bogota. 

1958 International Zoological Congress, London. Visits to Copenhagen and Hamburg, 

summer. 
1958-59 Organizing Committee for First International Congress of Oceanography, held in 
United Nations headquarters. New York, summer of 1959. 

1959 Elected honorary fellow. Zoological Society of India. 

1960 Field work and fish collecting during February, upper Rio Guaviare basin, near 

Sierra Macarena, Colombia, with General T. D. White. 
1960 Six-month study trip to Europe, visiting Hamburg, Copenhagen, Lund, Goteborg, 

Amsterdam, Leiden, Brussels, Frankfurt, Vienna, Lucerne, Paris, London. 
1963 Field work and fish collecting during February in Nicaragua; Managua area. Lake 

Nicaragua, Rio San Juan, with General T. D. White. 

1963 International Zoological Congress, Washington, D.C., August. 

1964 International Conference on Tropical Oceanography, Miami. Arranger and con- 

vener, section of zoogeography, November. 
1966 Marriage to Frances Edna Felin, Palo Alto, California. 



Vol. XXXVIII] WALFORD: HISTORY OF GEORGE S. MYERS 17 

1967 Primer Foro Internacional sobre Planificacion y Desarrollo Pesquero, Caracas, 

Venezuela, August. Followed by brief travel in eastern Venezuela and lower Rio 
Orinoco with Agustin Fernandez-Yepez. Visits to Trinidad, Panama and Puerto 
Rico en route. 

1969-70 Vice-President, Cactus and Succulent Society of California. 

1970 Statutory retirement on August 31 from faculty, Stanford, August. 

1970 Appointed Henry Bryant Bigelow Visiting Professor of Ichthyology, Harvard 
University. 



PROCEEDINGS 

OF THE 

CALIFORNIA ACADEMY OF SCIENCES 

FOURTH SERIES 



Festschrift for George Sprague Myers 



Vol. XXXVIII, No. 2, pp. 19-52. December 31, 1970 



ANNOTATED 

CHRONOLOGICAL BIBLIOGRAPHY 

OF THE PUBLICATIONS 

OF GEORGE SPRAGUE MYERS 

(to the end of 1969) 

This bibliography lists all known publications by George S. Myers, scientific 
and other, omitting only about two dozen ephemeral items such as newspaper 
articles. It has been compiled almost entirely from the mimeographed bibliogra- 
phies issued by IVIyers in 1950 and 1952 and his own card index of titles. Nearly 
all the bibliography and all the annotations are thus by Myers himself. The 
serial numbers of the papers have been used as annotations of mailed-out separata 
on Myers' address cards of colleagues, and have not been changed even though 
a few previously missing entries have now been inserted in their chronological 
places. 

To avoid confusion, Myers almost always gave separate numbers even to 
subsequent reprintings or translations of original contributions and to successive 
installments of serially published papers. In all such cases cross references are 
given. The total of numbered entries is thus greater than the number of original 
papers, but the usefulness of this system is obvious in the annotations. The last 
numbered entry is 593, but with ten interpolations added, the total is 603. 

Aside from formal taxonomic papers, the largest classes of publications are 
articles on aquarium fishes and book reviews. Myers became interested in 
ichthyology through aquarium fishes and retained his interest in them perma- 
nently. By 1930 he was already becoming the recognized authority on the identi- 

[19] 



20 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

fication and aquarium behavior of the smaller freshwater species from tropical 
America, Asia, and Africa. The number of Myers' publications in this field be- 
came especially large during the 26 months (June 19S2-August 1954) when he 
acted as editor of the Aquarium Journal. Although many of these papers were 
of a popular nature, many of them contain original observations on the taxonomy 
or behavior of the species concerned and could not be excluded even from a 
bibliography of scientific publications. 

In writing book reviews, Myers acted on a theory (reinforced, he says, by 
advice of the late Dr. Joseph Grinnell of the University of California) that care- 
fully done, critical reviews form a powerful instrument to weed out incompetence 
and raise the level of any scientific discipline. Myers obviously adhered to that 
theory, for his reviews are usually sharply critical and often embody original 
ideas or taxonomic views not expressed elsewhere. He has been known to say that 
he puts as much thought into reviews as into more formal contributions, and that 
some of his best writing has gone into them. A sampling shows this to be true; 
his reviews, editorials, and columns-of-comment give a broader view of his 
thought and critical abilities than do his formal papers. Many are worth reading 
at any time, especially by younger workers. 

There are also 49 papers dealing largely or wholly with zoological nomen- 
clature, ranging from his early attempts to modernize the names of aquarium 
fishes to such nomenclatural problems as the name of Cultcr (243), the family 
name of the characids (335), and some Neotropical frog names (560). 

Myers was asked by a former pupil if he could make a rapid analysis of his 
papers and came up with the following rough totals: (A) Papers of scientific 
interest on taxonomy, evolution, ecology, behavior, distribution, and nomencla- 
ture of fishes: 252. (B) Ditto of amphibians and reptiles: 51. (C) Historical 
zoogeography, chiefly of lower vertebrates: 8. (D) Formal book reviews: 104. 
(E) Editorials, comment, criticism: 11. (F) Taxonomic theory: 2. (G) Cura- 
torial, collecting, preserving: 5. (H) Popular articles on fishes, of little or no 
scientific interest: 95. (I) Botany: 2. (J) Station records: 1. (K) History 
of ichthyology: 2. (L) Biographies and obituaries: 14. (M) Translations and 
reprints of original Myers' papers: 32. (N) Continuations of serial articles with- 
out any change of title: 6. (O) Nonfish aquarium articles: 6. (P) Verse, 
allegory, etc.: 3. (Q) Unimportant notes and corrections: 9. Myers says that 
several papers published in the aquarium literature are included under category 
A, but only when they contributed significantly to knowledge of the fishes con- 
cerned. 

For the benefit of herpetologists, all papers of all categories that are im- 
mediately concerned with amphibians and reptiles are preceded by an asterisk 
(*) in the bibliography. 

The form of the entries is as follows: first the serial number, followed by 



Vol. XXXVIII] MYERS: BIBLIOGRAPHY 21 

the full title of the paper | untitled contributions have been given a title in 
brackets], the journal or other vehicle of publication, the volume number and 
issue number, and the pagination, followed by the precise date of publication so 
far as Myers knew it. Annotations by Myers are in brackets, usually at the end 
of the entry. The dates of papers, in cases where there is not even a month 
given, were known only as to year. Dates of publication of Copeia articles are 
the imprinted ones of the issues concerned. All dates given for the Stanford 
Ichthyologkal Bulletin papers, and for articles in the Aquarium Journal while 
Myers was editor (June 1952-August 1954) are the exact dates of mailing of 
the issues concerned. Where the imprinted year or month of a volume or article 
differs from the year placement or month placement in this bibliography (as 
with the proceedings of different Pacific Science Congresses) the date here given 
is the correct one for mailing of the volumes. Joint-authorship papers have the 
names of the authors given, in the original order, in parentheses immediately 
following the title. The only year in which no paper was published was 1968. 

Abbreviations of the names of journals frequently cited are as follows: 

AJ =z The Aquarium Journal. Published monthly by the San Francisco Aquarium Society, 
1928-1965. [Continued 1966-1967 as Ichtkyologica, The Aquarium Journal, published 
by TFH Publications, Inc., Jersey City, X. J.] Edited by G.S.M., 1952-1954. 

AL = Aquatic Life. Published by Joseph E. Bausman and edited by VV. A. Poyser in Phila- 
delphia, 1915-1922; subsequently published and edited for many years by August M. 
Roth in Baltimore. Philadelphia issues usually published promptly; Baltimore issues 
often lagged badly. 

MAN ^ American Museum Novitates. Published by The American Museum of Xatural 
History, New York City. Separate papers numbered consecutively; numbers separately 
paged. 

ANMG^ The Annals and Magazine of Xatural History. Published in London by Taj'lor 
and Francis, 1838-1966. Continued under the name Journal of Natural History. 

\P ^ Aquarium (Paris). Published in Paris, 1934-at least to 1936. Many articles from T.^ 
translated into French with the original TA colored plates and other figures. 

B\TK = Blatter fiir Aquarien- und Terrarienkunde. Published from 1890 until about 1937, 
first in Magdeburg and later in Stuttgart. Many important behavioral and taxonomic 
papers included. 

CO =: Copeia. Journal of the American Society of Ichthyologists and Herpetologists. Pub- 
lished at various places in the U.S.A., 1913-present. To 1930, all issues numbered con- 
secutively, with no volume number. Later, the year is the volume number (4 issues 
per year) . 

FC := The Fish Culturist. Published in Philadelphia by the Pennsylvania Fish Culturists' 
Association, 1921-present. 

H.'\B := The Home Aquarium Bulletin. Published at first in Newark, N. J., by the Newark 
Aquarium Society, later by a group in East Orange, N. J., 1931-to at least March, 1936. 
Carl L. Hubbs and Myron Gordon were among the associated editors. 

LSJ = Lingnan Science Journal. (Continuation of Lingnaam Agricultural Review.) Pub. 
lished by Lingnan University, Canton, China. Edited by Wm. E. Hoffman and Robert 
Cunningham Miller. 



22 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

PBSVV ^ Proceedings of the Biological Society oj Washington. Washington, D. C, 1880- 

present. 
SIQ= Stanford Ichthyological Bulletin. Edited by G. S. M. and M. H. Storey. Published 

by Natural History Museum of Stanford University (later Division of Systematic 

Biology), Stanford, Calif., 1938-1967. 
TA.^ The Aquarium. Published by Innes Publishing Co., Philadelphia, beginning in 1932. 

Edited by Wm. Thornton Innes. G. S. M. was an associate editor for many years. 
TFH = The Tropical Fish Hobbyist. Published by TFH Publishing Co., Jersey City, N. J., 

beginning in 1952. 
W ATK ^ Wochenschrift fiir Aquarien- und Terrarienkunde. Published weekly, 1904-1950, 

in Braunschweig. Much important behavioral and taxonomic information included. 

1920 

1. Phalloptychus januarius. AL, vol. 5, no. 7, p. 74. July. LMisidentification; species 

mentioned is spotted form of PhaUoceros caudimaculatus.] 

2. The red rivulus. AL, vol. 5, no. 7, pp. 79-80. July. LXanthic form of Rivulus 

urophthalmus.] 

3. Fundulus diaphanus. AL, vol. S, no. 8, p. 91. August. [The figure, supplied by the 

editor, is of F. heteroclitus.] 

4. Some fish suitable for home aquaria with suggestions concerning starting and main- 

taining an aquarium. 8vo, 16 pp., Hudson County Aquarium Society, Jersey City, 
N. J. September 10. [Circular distributed at 3rd annual exhibition of H.C.A.S.; 
pp. 2, 4, 6, and 8 are advertisements; author's name omitted by printer. Copies in 
the libraries of Stanford University and the U. S. National Museum.] 

5. The Mexican swordtail. AL, vol. 5, no. 11, pp. 122-123. November. 

1921 

6. The labyrinth fishes. L AL, vol. 6, no. 1, pp. 1-2. January. [Continued in nos. 13 

and 19.] 

7. Fundulus chrysotus. AL, vol. 6, no. 3, p. 18, September. 

8. The common sunfish. AL, vol. 6, no. 4, pp. 19-20. October-December. 

1922 

9. The black banded sunfish. Aquarium News, Ridgewood, N. Y., vol. 1, no. 5, p. 2. 

January 15. 

10. Chirodon arnoldi. AL, vol. 6, no. 5, p. 28. January-June. [The species mentioned is 

actually Astyanax mexicanus.] 

11. Planting aquaria. FC, vol. 2, no. 4, pp. 157-158. September. 

12. The aquarium and its denizens, being a brief exposition of the proper arrangement and 

maintenance of home aquaria, with a catalog of some of the fishes suitable for 
aquarium culture. 8vo, 22 pp. (cover is title-page) ; Hudson County Aquarium 
Society, Jersey City, N. J. September 8. [Copies in the libraries of Stanford Uni- 
versity, the British Museum, and the U. S. National Museum. See also no. 56.] 

13. The labyrinth fishes. IL AL, vol. 6, no. 6, pp. 33-34. July-October. [Continuation 

of 6. See also 19.1 

14. Interesting notes. FC, vol. 2, no. 6, pp. 172-173. November. [Notes on Fundulus 

gularis, Mesogonistius chaetodon, and Pterophyllum scalare; article includes six 
lines on p. 173, not well differentiated from other notes below.] 



Vol. XXXVIII] MYERS: BIBLIOGRAPHY 23 

15. Hudson County exhibition. AL, vol. 6, no. 7, p. 46. November. 

16. A true fish story. Swastika, Jersey City, vol. 1, no. 7, pp. 4-5. December 10. [Con- 

cerns the climbing perch. See 18 for note.] 

17. A recently described aquarium fish. CO, no. 113, p. 89. December 20. 

*18. The largest frog — the smallest frog. Swastika, Jersey City, vol. 1, no. 8, p. S. 
December 25. [A brief note on the size of Rana golialh and newly hatched 
Eleiitherodactylus . The Swastika was a student publication of the city high schools 
in Jersey City. Edited by Meyer Levin.] 

1923 

19. The labyrinth fishes. III. AL, vol. 6, no. 9, pp. 63-64. January. [Continuation of 6 

and 13. Never concluded.] 

20. The characins. FC, vol. 2, no. 8, pp. 186-187. January. 

21. A note on the fighting fish. FC, vol. 2, no. 10, pp. 202-203. March. [Note on the 

identification of aquarium examples of Betta; in the third paragraph, line four, the 
word "small" was transferred by printer's error from before "lake" in preceding line.] 

22. Correct names. FC, vol. 2, no. 11, pp. 210-212. April. [An attempt to modernize the 

scientific names of aquarium fishes. See also 24.] 

23. Aplocheilus chaperi. AL, vol. 7, no. 1, pp. 1-2, 12. May. 

24. Correct names. AL, vol. 7, no. 1, pp. 3-6. May. [A reprint of 22, with the addition 

of some name meanings.] 

25. Notes on the nomenclature of certain anabantid fishes and a new generic name pro- 

posed. CO, no. 118, pp. 62-63. May 20. 

26. Ctenobrycon spilurus. FC, vol. 2, no. 13, p. 226. June. 

27. On the subject of scavengers in the aquarium. FC, vol. 2, no. 12, pp. 228-229. June. 

28. A new limia from San Domingo. (John Treadwell Nichols and G. S. Myers.) AMN, 

no. 79, 2 pp. June 12. 

29. Hyphessobrycon anisitsi, a new fish for the aquarist. FC, vol. 3, no. 3, pp. 250-251. 

November. [Identification erroneous; species mentioned is Hemigrammus caudovit- 
tatus.] 

30. Further notes on anabantids. CO, no. 124, pp. 111-113. November 20. 

1924 

31. The labyrinth fishes. FC, vol. 3, no. 7, pp. 282-284. March. [Not part of the series: 

6, 13, and 19.] 

32. New genera of African poeciliid fishes. CO, no. 129, pp. 41-45. May 20. 

33. A new rivulus from Rio de Janeiro. ANMG, ser. 9, vol. 13, pp. 588-590. June. [R. 

dorni; holotype in British Museum. This species is a synonym of R. brasiliensis 
Valenciennes, according to G. S. M.] 

34. A new poeciliid fish from the Congo, with remarks on funduline genera. AMN, no. 

116, 9 pp. June 6. 

35. A new poeciliid fish of the genus Micropanchax from Ubangi. AMN, no. 112, 3 pp. 

June 24. 
*36. Amphibians and reptiles from Wilmington, N. C. CO, no. 131, pp. 59-62. June 30. 

37. On the existence of the Japanese killifish, FunduUchthys virescens. ANMG, ser. 9, vol. 

14, pp. 253-254. August. 

38. Mutanda ichthyologica. Neoborus Boulenger and Barbits rubyipinn'is Nichols and 



24 CALIFORNIA ACADEMY OF SCIENCES LProc. 4th Ser. 

Griscom. Revue Zoologie Africaine, Tervuren, Belgium, vol. 12, no. 3, p. 397. 
August 1. 

39. Liicania ommata in the aquarium. FC, vol. 4, no. 1, p. 314. September. 

40. A new poeciliid fish of the genus Rivitlus from British Guiana. AMN, no. 129, 2 pp. 

September 23. 

41. A new characin fish from Rio de Janeiro. FC, vol. 4, no. 3, pp. 330-331. November. 

[The legend for the figure was omitted by the printer; it is given correctly in 295. 
The two cotypes (syntypes) are now in the U. S. National Museum. See also 48.] 

42. On a small collection of fishes from Upper Burma. AMN, no. 150, 7 pp. November 13. 

43. The largest rivulus. CO, no. 135, p. 96. November 18. 

1925 

44. Concerning melanodimorphism in kilHfishes. CO, no. 137, pp. 105-107. January IS. 

LThe specimens of Platypoecilus coiichianiis mentioned were misidentified in the 
Field Museum collection; they were Mollienisia sphenops.] 

45. Description of a new catfish from Abyssinia. CO, no. 139, pp. 12-13. February IS. 

[See 520 for corrections.] 
45a. [Description of Rivulus rogoaguae.] (Nathan Everett Pearson and G. S. Myers.) 
In: Pearson, N. E., Fishes of the Rio Beni Basin (Indiana University Studies, 
"1924," vol. 11, no. 64), p. 51. February. 

46. Results of some recent studies on the American killifishes. FC, vol. 4, no. 8, pp. 

370-371. April. [Contains original diagnosis of Trigonectes strigabundus, n. gen. 
and n. sp. from the Rio Tocantins. Reprinted as 234.] 

47. Introduction of the European bitterling (Rhodeus) in New York and of the rudd 

(Scardinius) in New Jersey. CO, no. 140, pp. 20-21. April 14. 

48. Ein neuer Characinide von Rio de Janeiro. BATK, Jahrg. 36, no. 4, pp. 98-99, 

1 colored plate. April 15. (Translation of 41, omitting the original figure, but ac- 
companied by a colored plate by Curt Bessiger and by an article on breeding and 
care in aquaria, by Wilhelm Schreitmiiller.] 

49. Notropis cummingsi, a new minnow from Wilmington, North Carolina. AMN, no. 168, 

4 pp. April 2i. [Although named for a woman, Mrs. J. H. Cummings, the ending 
of cummingsi was intentional, the rationale being that Cummings was the name of 
her husband.] 

50. Concerning mollienisias. AL, vol. 9, no. 1, pp. i-A, 13-14. May. 

51. Astyanax fasciatus in the aquarium. AL, vol. 9, no. 3, p. 40. July. 

52. Description of a new cheirodontine characin from Rio de Janeiro. Annals of the 

Carnegie Museum, Pittsburgh, vol. 16, pp. 143-144, pi. 10. July 31. \Spinthcrobolus 
broccae, n. sp. =: Phoxinopsis typicus Regan.] 

53. Fishes changing sex. AL, vol. 9, no. 4, pp. 56-57. August. 

54. The blue characin, Coelurichthys microlepis. Aquarium Bulletin, St. Louis, vol. 2, no. 

3, pp. 3-4. September 1. [Reprinted as 95.] 

55. Labyrinth fishes for the aquarist. Aquarium Bulletin, St. Louis, vol. 2, no. 4, pp. 3, 6. 

October. 

56. The aquarium and its denizens, being a brief exposition of the proper arrangement 

and maintenance of home aquaria, with a catalog of some of the fishes suitable for 
aquarium culture. Second edition, revised and enlarged; 12mo, 45 pp., August M. 
Roth, PubHsher, Baltimore. November. [A revised edition of 12.] 

57. Tridentopsis pearsoni, a new pygidiid catfish from Bolivia. CO, no. 148, pp. 83-86. 

November 25. 



Vol. XXXVIII] MYERS: BIBLIOGRAPHY 25 

58. Fishes and human disease. FC, vol. 5, no. 4, pp. 2 7-29. December. [An account of 

the part played by fishes in malaria and yellow fever control. The reference to 
Astroblepiis {Ayges) is a lapsus; the genus intended is Pygidium.] 

1926 

59. Notes on anabantids. III. CO, no. 150, pp. 97-100. January 25. 

60. On the correct names of the tetra from Buenos Aires, the haplochilus from Madras, 

and the mouthbreeder. FC, vol. 5, no. 8, p. 61. April. 

61. Two new genera of African characin fishes. Revue Zoologie Africaine, Tervuren, 

Belgium, vol. 13, nos. 3-4, pp. 174-175. April 1. 

62. Die Xomenklatur der Labyrinthfische. BATK, Jahrg. 37, no. 8, pp. 190-193. April 30. 

63. Descriptions of a new characin fish and a new pygidiid catfish from the Amazon basin. 

CO, no. 156, pp. 150-152. July 20. [Proof not seen by author; there are several 
bad typographical errors which are corrected in 71.1 
*64. A synopsis for the identification of the amphibians and reptiles of Indiana. Proceed- 
ings of the Indiana Academy of Science, vol. 35 (1925), pp. 277-294. [Published in 
summer, 1926.] 

65. A cichlid fish that hangs its young on aquatic plants. AL, vol. 10, no. 4, pp. 60-61. 

August. 

66. Eine neue siidamerikanische Characinidenart der Gattung Pyn-huliiia. B.-\TK, Jahrg. 

37, no. 18, pp. 441-442. September 30. [The type locality, left indefinite in this 
paper, was later found to be Rosario, Argentina; syntypes now in U. S. X^ational 
Museum.] 

68. Alphabetical list of aquarium fishes, their breeding habits, care, etc. In: Innes, W. T., 

Goldfish varieties and tropical aquarium fishes, 9th ed., Philadelphia, pp. 264-283. 
October. [All of this chapter, save the introduction (pp. 264-265) and the conclud- 
ing remarks (pp. 283-284), is by G. S. Myers. This same list appeared in the later, 
cheaper edition of this book, called "The Complete Aquarium Book," published by 
Halcyon House, X. Y., in 1936.] 

69. Eine neue Characinidengattung der Unterfamilie Cheirodontinae aus Rio de Janeiro, 

Brasilien. BATK, Jahrg. 37, no. 24, pp. 566-567. December 30. [Syntypes of 
Rachoviscus crassiccps now in U. S. Xational Museum. This fish was probably not 
from the Baixada Flumenense but from farther inland. Perhaps equals Oligobrycoji?] 

1927 

70. An analysis of the genera of neotropical killifishes alHed to Rividus. AXMG, ser. 9, 

vol. 19, pp. 115-129. January. 

71. Xote [correcting typographical errors in no. 63.] CO, no. 158, pp. 167-168. January. 
*72. Rana areolata at Bloomington, Indiana. (Herman P. Wright and G. S. Myers.) CO, 

no. 159, pp. 173-175. January 11. [First description of eggs.] 

73. On the identity of the killifish Fundulus meeki Evermann with Fundulus lima \'aillant, 

CO, no. 160, p. 178. January 12. 

74. Puntis streeteri, a new cyprinoid fish from Borneo, and Cobitophh. a new genus of 

Bornean Cobitidae. AMX, no. 265, 4 pp. April 20. 

75. The status of the darter Richiella brevispina (Coker). CO, no. 163, pp. 39—13. June. 
*76. The differential characters of Bujo americanus and Bufo fowleri. CO, no. 163, pp. 

50-53. June. 
77. Descriptions of new South American fresh-water fishes collected by Dr. Carl Ternetz. 
Bulletin of the Museum of Comparative Zoology, Harvard, vol. 68, no. 3, pp. 107-135. 



26 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

July. [Types of Otothyris canalijerus now in U. S. National Museum, Paris, and 
London ; paratypes of Bunocephalus salathei in Washington. See also 163 for 
corrections.] 
78. A new genus of Brazilian characin fishes allied to Bivibranchia. (Carl H. Eigenmann 
and G. S. Myers.) Proc. National Academy of Sciences, Washington, vol. 13, no. 8, 
pp. 565-566. August. 
*79. Notes on Indiana amphibians and reptiles. Proceedings of the Indiana Academy of 
Science, vol. 36 (1926), pp. 337-340. September. 

80. Rasboras. FC, vol. 7, no. 2, pp. 175-177. October. 

1928 

81. Carl H. Eigenmann — Ichthyologist. Natural History, New York, vol. 28, no. 1, pp. 

98-101, portrait. 

82. The systematic position of the phallostcthid fishes, with diagnosis of a new genus from 

Siam. AMN, no. 295, 12 pp. February 1. 

83. The species of Piabucina inhabiting Colombia. CO, no. 166, pp. 4-5. March 23. 

84. Two new genera of fishes. CO, no. 166, pp. 7-8. March 23. 

85. Haplochilus cameronensis. AL, vol. 11, no. 12, p. 204. April. [A short note written 

in 1923 and published without author's knowledge in 1928. See 89.] 

86. The existence of cichlid fishes in Santo Domingo. CO, no. 167, pp. 33-36. June 28. 

87. The urostyle in larval characin fishes. CO, no. 167, pp. 36-37, June 28. 

88. New fresh-water fishes from Peru, Venezuela, and Brazil. ANMG, ser. 10, vol. 2, pp. 

83-90. July. [Two of the species are described by C. H. Eigenmann, and one jointly 
by Eigenmann and Myers.] 

89. "Haplochilus cameronensis." AL, vol. 12, no. 5, p. 94. September. [Correcting 85, 

above.] 

90. The characins. AL, vol. 12, no. 7, pp. 119-120, 122, 134-135. November. 

91. The characins. AL, vol. 12, no. 8, p. 152. December. [Continuation of 90.] 

1929 

92. The happy family tank. FC, vol. 8, no. 5, pp. 51-52. January. [On "community 

aquaria."] 

93. The history of the veiltail fighting fish. AL, vol. 12, no. 10, pp. 195-199. February. 

[History of the original introduction of long tailed cultivated Betta splendens into 
the United States.] 

94. Cranial differences in the African characin fishes of the genera Alestes and Brycinus, 

with notes on the arrangement of related genera. AMN, no. 342, 7 pp. March 2. 

95. The blue characin, Mimagoniates microlepis. FC, vol. 8, no. 8, pp. 92-93. April. [A 

reprint of 54, above, with emendations.] 

96. Mutanda icthyologica. II. Heringia vs. Rhiiiosardinia (Clupeidae), Medipellona vs. 

Chirocentrodon (Clupeidae), and Entonanthias vs. Mirolabrichthys (Anthiidae). 
CO, no. 170, pp. 1-2. April 30. 

97. A note on the Formosan homalopterid fish, Crossostoma lacustre Steindachner. CO, 

no. 170, p. 2. April 30. 
*98. Notes on the names of the spring peeper, the carpenter frog, and Ancides aeneiis. CO, 
no. 170, pp. 22-23. April 30. [See 121 for correction.] 
99. On curimatid characin fishes having an incomplete lateral line, with a note on the 
peculiar sexual dimorphism of Curimatopsis tnacrolepis. ANMG, ser. 10, vol. 3, 
pp. 618-621. June. 



Vol. XXXVIII] MYERS: BIBLIOGRAPHY 27 

100. Notes on soles related to Achiriis. CO, no. 171, pp. 36-38. June 28. 

101. The American Characidae. [Part 5] (Carl H. Eigenmann and G. S. Myers.) Memoirs 

of the Museum of Comparative Zoology, Harvard, vol. 43, part 5, pp. 429-558, 11 
pis. September. [The supplement, pp. 516-550, is nearly all by G.S.M; the rest of 
the text is nearly all by Eigenmann.] 

102. Our aquarium fishes. I. The mouthbreeder. AJ, vol. 2, no. 8, p. 31. October 3. 

1930 

*103. Amphibians and reptiles observed in the Palisades Interstate Park, New York and 

New Jersey. CO, no. 173, pp. 99-103. January 16. 
*104. Notes on some amphibians in western North America. PBSW, vol. 43, pp. 55-64. 

March 12. 
105. Fishes from the upper Rio Meta basin, Colombia. PBSW, vol. 43, pp. 65-71. March 12. 
*106. The status of the southern CaHfornia toad, Btifo californicus (Camp). PBSW, vol. 43, 

pp. 73-77. March 12. 

107. On the occurrence and habits of ocean sunfish {Mola mola) in Monterey Bay, Cali- 

fornia. (G. S. Myers and Joseph Howe Wales.) CO, 1930, no. 1, pp. 11-12. April 30. 

108. The killifish of San Ignacio and the stickleback of San Ramon, Lower California. 

Proceedings of the California Academy of Sciences, ser. 4, vol. 19, no. 9, pp. 95-104. 
July IS. 

109. Ptychidio jordani, an unusual new cyprinoid fish from Formosa. CO, 1930, no. 4, 

pp. 110-113. December 31. [Type later found to represent a chance introduction in 
Formosa, originating from the pondfish-fry export industry centering at Wuchow. 
Genus and species are endemic to the Si Kiang (West River) system, near to and 
above Wuchow, Kwangsi, China.] 

110. [Review of] Publications of the University of Oklahoma Biological Survey, vol. 1. 

By A. Richards, C. L. Hubbs, and A. I. Ortenburger. CO, no. 4, pp. 159-160. 
December 31. 

111. [Review, with critical comments, of] Hall. H. M., and Clements, F. E., The 

phylogenetic method in taxonomy. Micropaleontology Bulletin, Stanford Univer- 
sity, vol. 2, no. 3, pp. 55-58. December 31. 

1931 

112. Eigenmann, Carl H. lu: Dictionary of American Biography, vol. 6, pp. 62-63. 

Charles Scribner's Sons, New York. 

113. Killifishes in Hispaniola. FC, vol. 10, no. 6, pp. 103-104. February. 

114. Ichthyological reminiscences of a trip east. AJ, vol. 4, no. 2, p. 9. February. 

115. Ichthyological reminiscences of a trip east (continued). .'\J, vol. 4, no. 3, pp. 14-15. 

March 5. 

116. Ichthyological reminiscences of a trip east (concluded). AJ, vol. 4, no. 4, pp. 20-21. 

April 2. 

117. The primary groups of oviparous cyprinodont fishes. Stanford University PubHcations, 

University Series, Biological Sciences, vol. 6, no. 3, pp. 241-254. [Copies first 
mailed April 7.] 
*118. Ascaphus tritei in Humboldt County, California, with a note on the habits of the 
tadpole. CO, 1931, no. 2, pp. 56-57. July 20. 
119. Fishes from southeastern China and Hainan. (Albert W. Herre and G. S. Myers.) 
LSJ, vol. 10, no. 2-3, pp. 233-254. .'\ugust. [Key to Asiatic genera of Clupeidae 
by G.S.M. alone. See 126 for correction.] 



28 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

120. On the fishes described by Roller from Hainan in 1926 and 1927. LSJ, vol. 10, nos. 
2-3, pp. 2SS-262. August. 
*121. The original descriptions of Bufo fowleri and Biifo americanus. CO, 1931, no. 3, pp. 
94-96. October 30. 

122. Poeciliid fishes of the genus Mollienisia in Hispaniola, with notice of a new limia from 
the Samana Peninsula. AMN, no. 503, 2 pp. November 9. 

123. On the identity of Ophicephahis and Channa, two genera of labyrinth fishes. (G. S. 

Myers and Leo Shapovalov.) Peking Nat. Hist. Bull., Peiping, China, vol. 6 (1931- 
32), pt. 2, pp. 33-37. November. [Vol. 6, part 2 has usually been considered to 
have been published in 1932. Copies of vol. 6, part 2 reached regular subscribers in 
California in December 1931, and publication in China must have been in November 
or earlier.] 

1932 

124. Gambusias in the aquarium. HAB, vol. 2, no. 1, pp. 2-5. March. [See also 150.1 

125. The osteoglossid fish Sclcropages in the Malay Peninsula. CO, 1932, no. 1, p. 30. 

April 12. 

126. Nealosa Herre and Myers equals Konosirus Jordan and Snyder. CO, 1932, no. 1, 

p. 30. April 12. 

127. A new name for a Melanesian pseudochromid fish confused with Nesiotes piirpurascens 

de Vis. CO, 1932, no. 1, p. 30. April 30. 
*128. A neglected description of a Mexican garter-snake, Thamnophis stejnegeri McLain. 
CO, 1932, no. 1, p. 35. April 12. 

129. [Review of] Osborn, H. F. Cope: Master Naturalist, and Biographical memoir of 

Edward Drinker Cope. CO, 1932, no. 1, pp. 39^1. April 12. 

130. Danio analipunctatus identified as Brachydanio nigrofasciatus. TA, vol. 1, no. 2, 

p. 54. June. [Boulenger's holographic description of Danio analipunctatus, which 
he sent to J. P. Arnold and which was published by Arnold in BATR, was sent 
to G.S.M. by Arnold, and is now in Library of U. S. National Museum.] 

131. A note on the two Chinese paradise fishes. HAB, vol. 2, no. 5, p. 9. July. [Pre- 

liminary synopsis of 137.] 

132. Some new aquarium fishes from Panama. TA, vol. 1, no. 3, pp. 68-69, 82. July. 

133. Some notes on the characin, Astyanax mexicamis, in Texas. AL, vol. 16, no. 3, 

pp. 97-98. July. 

134. A new gonostomatid fish, Neophos nexili.s, from the Philippines. CO, 1932, no. 2, 

pp. 61-62. July 1. 

135. A new whitefish, Prosopium snyderi, from Crescent Lake, Washington. CO, 1932, no. 2, 

pp. 62-64. July 1. 

136. Fundulus chrysotus, geschechte Abart. WATR, Jahrg. 29, no. 29, pp. 450-451. July 19. 

[Notes on black-spotted, melanic or melanodimorphic specimens of Fundulus, 
Gambusia, and Mollienisia. See also 44.] 

137. The two Chinese labyrinth fishes of the genus Macropodus. LSJ, vol. 11, no. 3, pp. 

385-403, pis. 6-7. July 22. [A taxonomic revision of the genus. Proof not seen 
by author; many typographical and editorial errors, especially in explanation of 
plates. See also 131; see 576 for additional information and another synonym.] 

138. A rare deep-sea scombrid fish, Xenogramma carinatum Waite, on the coast of southern 

California. Trans. San Diego Society of Natural History, vol. 7, no. 11, pp. 111-117, 



Vol. XXXVIII] MYERS: BIBLIOGRAPHY 29 

pi. 7. July 28. [First American record. First synonymization of several nominnl 
species. Species later known as Lepidocybium jlavobrunneum (A. Smith) 1849.] 

139. Dangers in identifications. TA, vol. 1, no. 4, pp. 94-97, 110. August. 

140. Pterophyllum, king of aquarium fishes. TA, vol. 1, no. 5, pp. 115-118, 140-141. 

September. [A systematic review of the 3 species.] 

141. A new genus of funduline cyprinodont fishes from the Orinoco Basin, Venezuela. 

PBSW, vol. 45, pp. 159-162. September 27. 

142. Recent importations — Stethaprion innesi and Mylossoma aiirem from the Amazon. 

TA, vol. 1, no. 6, pp. 149-150, 171. October. [This will probably have to stand as 
the original description of 5. innesi, which was formally described in 148. Re- 
printed as 262.] 

143. A native fish, Notropis lutrensis, in the aquarium. AJ, vol. 5, no. 8, pp. 45-46. 

October 6. 

144. Notes on Colombian fresh-water fishes, with description of a new astroblepus. CO, 

1932, no. 3, pp. 137-138. October 7. 

1933 

145. Pachypanchax, a new genus of cyprinodont fishes from the Seychelles Islands and 

Madagascar. AMN, no. 592, 1 p. January 23. 

146. A new genus of Chinese fresh-water serranid fishes. Hong Kong Naturalist, vol. 4, 

no. 1, p. 76. April. [Proposing the new genus Acroperca, which was antedated by 
Coreosiniperca Fang and Chong 1932. The difference in date was about three 
months.] 
*147. Two records of the leatherback turtle on the California coast. CO, 1933, no. 1, p. 44. 
April 3. 

148. Description of a new characid fish of the genus Stethaprion from the Lower Amazon. 

ANMG, ser. 10, vol. 11, pp. 604-605. May. [See 142.] 

149. Stevardia albipinnis? HAB, vol. 3, no. 4, p. 11. June. 

150. Gambusen im Aquarium. WATK, Jahrg. 30, no. 26, pp. 401-403. June 27. [Transla- 

tion of 124.] 

151. [Review of] Hora, S. L., Classification, bionomics and evolution of homalopterid 

fishes. CO, 1933, no. 2, p. 109. July 20. 

152. The classification of the African cyprinodont fishes with a discussion of the geo- 

graphical distribution of the Cyprinodontidae of the world. Stanford Univ. Bull., 
ser. 5, no. 158 (Abstracts of dissertations, vol. 8, 1932-33), pp. 10-12. July 31. 
[Brief abstract of doctorate thesis, which was more usefully abstracted as 157 and 
214.] 

153. Anent Mr. Schoenfeld on scientific names. HAB, vol. 3, no. 6, pp. 6-11. August. 

154. New importations — "Jack Dempsey" unmasked. Aquarium, Philadelphia, vol. 2, no. 6, 

pp. 141-142. October. [Reidentifies fish previously known to aquarists as Cichla- 
soma nigrofasciatum — the "Jack Dempsey" — as C. biocellatum.'] 

155. Note on the breeding habits of Corynopoma. FC, vol. 13, no. 3, p. 61. November. 

156. New importations. — Leopard Corydoras. TA, vol. 2, no. 8, pp. 188-189. December. 

[Contains first diagnosis of Corydoras leopardus, which is compared to C. jidii and 
C. trilineatus. The species is formally described in 178. Translated as 166.] 

157. The genera of Indo-Malayan and African cyprinodont fishes related to Panchax and 

Nothobranchiiis. CO, 1933, no. 4, pp. 180-185. December 27. [See also 180.] 



30 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

1934 

158. The identification of aquarium fishes related to Metynnis and Serrasalmus. FC, vol. 

13, no. 5, pp. 120-122. January. [Generic identification of live specimens.] 

159. New importations — The black-winged flying characin. (Carnegiella marthae Myers.) 

TA, vol. 2, no. 9, pp. 217-218. January. fFirst account of life colors and habits, 
first record from Amazon basin, and first photograph of living specimens. See also 
186 and 289, and comments in 345.] 

160. [Radio interview on fish work in the National Museum.] AL, vol. 17, no. 9, pp. 239- 

240. January. 

161. [Review of] Coates, W., Tropical fishes for a private aquarium. FC, vol. 13, no. 6, 

p. 154. February. 

162. Reports on the collections obtained by the first Johnson-Smithsonian Deep-sea Ex- 

pedition to the Puerto Rican Deep. Three new deep-water fishes from the West 
Indies. Smithsonian Miscellaneous Collections, vol. 91, no. 9, 12 pp., 1 pi. April 2. 

163. Corrections of the type localities of Metzia mesembrina, a Formosan cyprinid, and of 

Othonocheirodus eigenmanni, a Peruvian characin. CO, 1934, no. 1, p. 43. April 24. 

164. A new name for the Alaskan cottoid fish Ulca marmorata (Bean). CO, 1934, no. 1, 

p. 44. April 24. 

165. [Review of] Regan, C. T., and Trewavas, E., Deep-sea angler fishes (Ceratioidea). 

CO, 1934, no. 1, pp. 54-55. April 24. 

166. Le Corydoras Leopard. AP, no. 5, p. 77. May. [A re-edited translation of 156.] 

167. Barbiis partipentazona Fowler. TA, vol. 3, no. 4, p. 83. August. [See also 182.] 

168. Our downtrodden helper, the snail. AL, vol. 18, no. 4, p. 74-76, 93. August. [Snails 

in aquaria.] 

169. Judging fish shows. TA, vol. 3, no. 5, pp. 103-106. September. 

170. Gnathocharax steindachneri, a new characin for the aquarium. HAB, vol. 4, no. 7, pp. 

5, 29-30. September. [Records species from the Orinoco (Caicara) and from 
British Guiana (Rockstone) for the first time; also records Monocirrhus poly- 
acanthus from Rockstone, British Guiana.] 

171. Correct nomenclature. AL, vol. 18, no. 6, pp. 139-141. October. [Nomenclature of 

aquarium fishes.] 

172. [Review of] Lederer, N., Tropical fish and their care. CO, 1934, no. 3, p. 143. 

October 31. 

173. Ueber den Namen des Zwergdrachenflossers, Corynopoma riisei Gill {=:Stevardia 

albipinnis Gill). WATK, Jahrg. 31, no. 48, pp. 755-756. November 27. 

174. [Review of] Herre, A. W. C. T., Notes on fishes in the zoological museum of Stan- 

ford University, 1: The fishes of the Herre Philippine Expedition of 1931. CO, 
1934, no. 4, pp. 196-197. December 31. 

1935 

175. Cichlasoma biocellatum. TA, vol. 3, no. 9, p. 196. January. 

176. The mouth-breeding fighting fish, Betta brederi. TA, vol. 3, no. 9, p. 210. January. 

[This must stand as the original description of B. brederi, which is formally des- 
cribed in 179. Translated as 181 and 191.] 

177. A new phallostethid fish from Palawan. PBSW, vol. 48, pp. 5-6. February. [Pro- 

poses new suborder Phallostethoidea.] 



Vol. XXXVIII] MYERS: BIBLIOGRAPHY 31 

178. Four new fresh-water fishes from Brazil, Venezuela and Paraguay. PBSVV, vol. 48, pp. 

7-13. February 6. [One new species, Corydoras leopardus, was first diagnosed in 
156.] 

179. A new anabantid fish of the genus Betta from Johore. PBSW, vol. 48, pp. 25-26. 

February 6. [First diagnosed in 176.] 

180. The genera of Indo-Malayan and African cyprinodont fishes allied to Panchax and 

N othobranchius . Canadian Aquaria, London, Ontario, vol. 3, no. 3, pp. 42-45. 
March. [Unauthorized reprint of part of 157.] 

181. Der maulbrutende Kampffisch Betta brederi. WATK, Jahrg. il, no. 20, p. 307. May 

14. [Translation of no. 176. See also 191.] 

182. Barbus partipeniazona, Fowler. AP, no. 19, p. 111. July. [Translation of 167.] 

183. [Review of] Schreitm tiller, W., Leitfaden zur Pflege und Zucht von einheimischen 

und fremlandischen Zierfischen, Seetieren, Schnecken und Wasserpflanzen nebst 
einem Anhang fiber Trocken- und Kunstfischfutterarten; 3rd edition. CO, 1935, no. 
2, p. 107. July 18. 

184. [Review of] Holly, M.; Meinken, H.; and Rachow, A., Die Aquarienfische in Wort 

und Bild. CO, 1935, no. 2, p. 107. July 16. 

185. Fishes of the upper Potomac. In: Shosteck, Robert, The Potomac trail book (128 pp., 

12mo; published by the Washington Post, Washington, D. C), pp. 90-95. [Fish 
chapter edited by Shosteck and proof not seen by G.S.M. before pubhcation. Para- 
graphs on the eel and the goldfish on pp. 94-95 were added by Shosteck. The term 
"upper Potomac" was insisted on by Shosteck ; the paper concerns the fishes of the 
region above the estuary but below Great Falls. The book was published (placed on 
sale) on October 6. Copies in libraries of U.S. National Museum, Carl L. Hubbs, 
and G.S.M.] 

186. Carnegiella marthae Myers. AP, no. 2i, pp. 173-174. November. [Translation of 159. 

See also 289.] 

187. An annotated list of the cyprinodont fishes of Hispaniola, with descriptions of two 

new species. Zoologica, New York, vol. 10, no. 3, pp. 301-316. November 29. 

188. Reports on the collections obtained by the first Johnson-Smithsonian Deep-sea Ex- 

pedition to the Puerto-Rican Deep. A new genus of opisthognathid fishes. Smith- 
sonian Miscellaneous Collections, vol. 91, no. 23, 5 pp. December 24. [Contains 
notes on Owstoniidae, Macrurocyttus, etc.] 

189. [Review of] Two new state ichthyological papers [Greene, C. W., the distribution of 

Wisconsin fishes, and O'Donnell, D. J., Annotated list of the fishes of Illinois]. CO, 
1935, no. 4, pp. 196-197. December 31. [In this was attempted the only pubhshed 
history and evaluation of the unfortunate check-list of North American fishes by 
Jordan, Evermann, and Clark.] 

1936 

190. [The use of fishes in mosquito control.] /;;: Sweetman, H. L., the biological control 

of insects (Comstock Publishing Co., Ithaca, N. Y.), pp. 318-325. [This chapter, 
although not credited to G.S.M. as author, was written by him and is printed from 
the MS. he supplied to Sweetman, with the change of only a very few words. Most 
of the bibliography and references given in the MS., however, were omitted by 
Sweetman.] 

191. Le Betta brederi. AP, no. 25, p. 15. January. [A translation of 176, with additional 

notes on habits and breeding. See also 181.] 



U CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

192. Note on Rhamphichthys cingulatus Brind. Aquarium News, Rochester, N. Y., vol. 3, 

no. 5, p. 68. January. 

193. On the Indo-Australian fishes of the genus Scatophagus, with description of a new 

genus, Selenotoca. PBSW, vol. 49, pp. 83-85. July 3. 

194. A new characid fish of the genus Hyphessobrycon from the Peruvian Amazon. PBSW, 

vol. 49, pp. 97-98. July 3. 

195. A third record of the albulid fish Dixonina nemoptera Fowler, with notes on an 

albulid from the Eocene of Maryland. CO, 1936, no. 2, pp. 83-85. July 31. 

196. A note on the stephanoberycid fishes. CO, 1936, no. 2, p. 118. July 31. [Resuscitates 

Acanthochaenus liietkenii G'lW — Stephanoberyx gilii.] 

197. [Review of] Herre, A. W. C. T., The fishes of the Crane Pacific Expedition. CO, 1936, 

no. 2, pp. 128-129. July 31. [Notes that Disparkhthys is probably not a genus of 
eels and may be near the blennies, and that Alepideleotris equals Eleotrica.^ 

198. A new genus of gymnotid eels from the Peruvian Amazon. PBSW, vol. 49, pp. 115- 

116. August 22. lOedemognathus exodon; also synonymizes Tateichthys duidae 
LaMonte with Steatogenys elegans (Steindachner).] 

199. A new polynemid fish collected in the Sadong River, Sarawak, by Dr. William T. 

Hornaday, with notes on the genera of Polynemidae. Journal of the Washington 
Academy of Science, vol. 26, no. 9, pp. 376-382. September 15. [Revision of 
genera of Polynemidae.] 

200. Report on the fishes collected by H. C. Raven in Lake Tanganyika in 1920. Proceed- 

ings of the United States National Museum, vol. 84, no. 2998, pp. 1-15, pi. 1. 
September 24. 

201. Description of a new blennioid fish of the genus Acanthemblemaria from the Pacific 

coast of Panama. (G. S. Myers and Earl D. Reid.) Allan Hancock Pacific Expedi- 
tions, vol. 2, no. 2, pp. 7-10. December. 

202. [Note on the identity of the dead fish in the Tidal Basin at Washington, D. C] 

PBSW, vol. 49, p. viii. [Oral communication ; no title. It concerned thousands of 
adults of Opjsthonema oglinum floating dead in the Tidal Basin after thawing of 
the heavy ice in spring, 1936. This section of the Proceedings was published at the 
end of the year.] 

1937 

203. Notes on phallostethid fishes. Proceedings of the United States National Museum, 

vol. 84, no. 3007, pp. 134-143. January 6. 

204. The deep-sea zeomorph fishes of the family Grammicolepidae. Proceedings of the 

United States National Museum, vol. 84, no. 3008, pp. 145-156, 3 pis. January 18. 

205. [A] possible method of evolution of oral brooding habits in the cichlid fishes. AJ, vol. 

10, no. 4, pp. 4-6. April. [Printer omitted the "A" in the title. Reprinted in facsimile, 
except for title, as 218. See also 223.] 

206. A contribution to the ichthyology of the Malay Peninsula, Part II. Fresh-water 

fishes. (Albert W. C. T. Herre and George S. Myers.) Bull. Raffles Mus., Singa- 
pore, no. 13, pp. 53-75, pis. 5-7. August. [The first part of this paper, on marine 
fishes, and the general introduction, are by Herre alone, although the title page 
would lead one to think otherwise.] 

207. [Review of] Monographs on the fishes of the Iberian Peninsula and Madeira. IBuen, 

F. de, Catalogo de los peces Ibericos; Lozano Rey, L., Los peces fluviales de 
Espana; Nobre, A., Fauna marinha de Portugal, I, Vertebrados; Noronha, A. C. de, 



Vol. XXXVIII] MYERS: BIBLIOGRAPHY 33 

and Saimento, A. A., Os peixes dos mares do Madeira.] CO, 1937, no. 4, pp. 
239-240. December 31. 

208. [Review of] Three new Asiatic check lists LSuvatti, C, Index to fishes of Siam ; Roxas, 

H. A., and Martin, C, A check Hst of Philippine fishes; Mori, T., and Uchida, K., 
A revised catalogue of the fishes of Korea]. CO, 1937, no. 4, pp. 241-242. Decem- 
ber 31. 

1938 

209. [Review of] Hubbs, C. L., and Trautman, M. B., A revision of the lamprey genus 

Ichthyomyzon. CO, 1938, no. 1, p. 51. March 31. 

210. Foreword. SIB, vol. 1, no. 1, pp. 1-2. June 22. [Explains editorial policy and aims 

of the Bulletin.] 
*211. Hydromantes platycephahis in Sonora Pass, CaHfornia. CO, 1938, no. 2, p. 91, June 30. 

212. Notes on Ansorgia, Clarisilurus, Wallago, and Ceratoglanis, four genera of African 

and Indo-Malayan catfishes. CO, 1938, no. 2, p. 98. June 30. [See 322 for correc- 
tion.] 

213. Fresh-water fishes and West Indian zoogeography. Annual Report Smithsonian In- 

stitution for 1937, pubhcation 3465, pp. 339-364, 3 pis. [volume appeared late in 
the summer of 1938.] 

214. Studies on the genera of cyprinodont fishes. XIV. Aplocheilkhthys and its relatives 

in Africa. CO, 1938, no. 3, pp. 136-143. September 24. [Numbers 32, 34, 37, 46, 70, 
84, 117, 141, 145, 157, 178, 187, and 200 are taken to be the first 13 papers in this 
series.] 

215. [Review of] Wells, L. A., Tropical aquariums, plants and fishes. CO, 1938, no. 3, 

p. 152. September 24. 

216. Harvest of the sea. Stanford Illustrated Review, vol. 40, no. 2, pp. 20-21, 25-26. 

October. [Originally given as a radio address.] 

217. [Review of] Costen, H. E. T., Beneath the surface, the cycle of river life. CO, 1938, 

no. 4, p. 208. December 10. 

1939 

218. A possible method of evolution of oral brooding habits in cichlid fishes. SIB, vol. 1, 

no. 3, pp. 85-87. February 3. [A reprint, in photographic facsimile except for the 
title, of 205. See also 223.] 

219. Notes on the labrid genus Lienardella. SIB, vol. 1, no. 3, pp. 87-88. February 3. 

220. On the Brazilian characid fish Notropocharax difficilis Marini, Nichols and La Monte. 

SIB, vol. 1, no. 3, p. 8. February 3. 

221. [Review of] Norman, J. R., Discovery reports, coast fishes. CO, 1939, no. 1, p. 61. 

March 9. 

222. A new owstoniid fish from deep water off the Philippines. PBSW, vol. 52, pp. 19-20. 

March 11. 

223. Mouthbreeding in cichlid fishes. Aquarist and Pond-keeper, London, vol. 9, no. 3, pp. 

90-91,94. May. [Reprint of 205. See also 218.] 

224. Hesperomyriis fryi, a new genus and species of echelid eels from California. (G. S. 

Myers and Margaret Hamilton Storey.) SIB, vol. 1, no. 4, pp. 156-159. May 24. 

225. The possible identity of the Congo fish Teleogranima with the cichlid genus 

Leptolamprologus. SIB, vol. 1, no. 4, p. 160. May 24. 



34 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

226. A living coelacanth fish. CO, 1939, no. 2, p. 124. July 12. [A note on the discovery 

of Latimeria chalumnae in South Africa, with references to accounts in popular 
magazines.] 

227. [Review of] Watson, D. M. S., The acanthodian fishes. CO, 1939, no. 3, p. 178. 

September 9. 

228. [Review of] Tchernavin, V., Changes in the salmon skull. CO, 1939, no. 3, p. 178. 

September 9. 

229. [Obituary of Alipio de Miranda-Ribeiro, 1874-1939.] CO, 1939, no. 3, p. 184. Septem- 

ber 9. [See also 535.] 

230. [Review of] Moy-Thomas, J. A., Palaeozoic fishes. CO, 1939, no. 4, p. 239. Decem- 

ber 26. 

231. [Review of] Clements, F. E., and Shelford, V. E., Bio-ecology. (Lionel Albert Wal- 

ford and G. S. Myers.) CO, 1939, no. 4, p. 240. December 26. 

1940 

232. Suppression of Acaropsis and Chalcimis, two preoccupied names of South American 

fresh-water fishes. SIB, vol. 1, no. 5, p. 170. February 7. 

233. On the use of the generic name Barbus in ichthyology and ornithology. SIB, vol. 1, 

no. 5, p. 170. February 7. [Shows that "Barbus" of Cuvier is a plural vernacular 
term for a bird group, and does not preoccupy the fish name Barbus.] 

234. Results of some recent studies on the American killifi.shes. SIB, vol. 1, no. 5, pp. 171- 

172. February 7. [A photographic facsimile of 46.] 

235. Zoological nomenclature. Nature, London, vol. 145, no. 3668, pp. 264-265. February 

17. [The word "many" in the MS. was misprinted "any" in fourth from last line 
of the last paragraph.] 

236. [Review of] Neave, S. A., Nomenclator zoologicus. CO, 1939, no. 1, p. 56. March 30. 

237. [Review of] Kuhne, E. R., A guide to the fishes of Tennessee and the mid-south. 

CO, 1939, no. 1, p. 58. March 30. 

238. [Remarks on resolution regarding dams and migratory fishes.] SIB, vol. 1, no. 6, 

p. 209. May 3. 

239. Cope as an ichthyologist. CO, 1940, no. 2, pp. 76-78, 1 pi. July 28. [In Edward 

Drinker Cope Centenary Number.] 

240. A note on Monognathus. CO, 1940, no. 2, p. 141. July 28. [Selects a type species 

for Monognathus and proposes new genus Phasmatostoma for other species.] 

241. The probable identity of Sphyraena chrysotaenia from the Red Sea and Arabia with 

S. aureoflamma from the Philippines, with notes on Naso vomer and A^. lopezi. 
CO, 1940, no. 2, p. 143. July 28. 

242. [Review of] Outright, P. R., The great naturalists explore South America. CO, 1940, 

no. 2, p. 145. July 28. 

243. The nomenclatural status of the Asiatic fish genus Culler. CO, 1940, no. 3, pp. 199- 

201. November 14. [See 516.] 

244. A note on the status of the generic name Corydoras. SIB, vol. 2, no. 1, pp. 11-12. 

December 23. [Inserted as part of paper by William Gosline on the Callichthyidae.] 

245. Suppression of some preoccupied generic names of fishes (Kessleria, Entomolepis, 

Pterodiscus, and Nesiotes), with a note on Pterophyllum. SIB, vol. 2, no. 1, pp. 
33-36. December 23. [The cichlid generic name Pterophyllum is not preoccupied.] 

246. An American cyprinodont fish, Jordanella floridae, reported from Borneo, with notes 

on the possible widespread introduction of foreign aquarium fishes. CO, 1940, no. 4, 



Vol. XXXVIII] MYERS: BIBLIOGRAPHY 35 

pp. 267-268. December 27. [First warning of the now worldwide introductions of 
tropical freshwater aquarium fishes in warmer regions.! 

247. The neotropical anchovies of the genus Amplova. Proceedings of the California 

Academy of Sciences, 4th ser., vol. 2i, no. 29, pp. 437-442. December 31. [This 
paper, written in 1926, follows Eigenmann's system of giving total lengths (including 
caudal) of the specimens, but the length of the holotype of Amplova alleni, as given, 
is the standard length. The holotype was selected by Hildebrand from among what 
G.S.M. had intended to be syntypes of the species.] 

1941 

248. Suppression of Lissockilus in favor of Acrossocheilns for a genus of Asiatic cyprinid 

fishes, with notes on its classification. CO, 1941, no. 1, pp. 42-44. March 25. 

249. [Review of] Huxley, J., The new systematics. CO, 1941, no. 1, p. 61. March 25. 

250. The fish fauna of the Pacific Ocean, with especial reference to zoogeographical regions 

and distribution as they affect the international aspects of the fisheries. Proceedings, 
Sixth Pacific Science Congress, vol. 3, pp. 201-210. [Vol. 3 was issued in April 1941.] 

251. The work and program of the Natural History Museum of Stanford University in 

fisheries and general ichthyology. Proceedings, Sixth Pacific Science Congress, vol. 
3, pp. 413-415. April. [See note under 250.] 

252. Four new genera and ten new species of eels from the Pacific coast of tropical America. 

(G. S. Myers and Charles Barkley Wade.) Allan Hancock Pacific Expeditions, vol. 
9, no. 4, pp. 65-111, pis. 7-16. June 25. 

253. [Review of] Schuchert, C, and Le Vene, C. M., O. C. Marsh, pioneer in paleontology. 

CO, 1941, no. 2, p. 121. July 8. 

254. [Review of] Sherborn, C. D., Where is the collection? An account of the various 

natural history collections which have come under the notice of the compiler. CO, 
1941, no. 2, p. 122. July 8. 
*255. [Review of] Fitch, H. S., A biogeographical study of the ordinoides Artenkreis of 
garter snakes (genus Thamnophis) . CO, 1941, no. 2, pp. 122-123. July 8. 

256. [Review of] Phillips, W. J., The fishes of New Zealand. CO, 1941, no. 3, p. 187. 

September 30. 

257. [Review of] Berg, L. S., Classification of fishes, both recent and fossil. CO, 1941, 

no. 4, pp. 2 74-2 75. November 21. 

258. [Review of] Hubbs, C. L., and Lagler, K. F., Guide to the fishes of the Great Lakes 

and tributary waters. CO, 1941, no. 4, p. 275. November 21. 

259. [Review of] Parker, T. J., and Haswell, W. A., A text-book of zoology, 6th edition. 

CO, 1941, no. 4, pp. 275-276. November 21. [Comments on the decline of 
morphological zoology.] 

260. [Review of] Norris, H. W., The plagiostome hypophysis, general morphology and 

types of structure. CO, 1941, no. 4, p. 277. November 21. 

261. A new name for Taenionema, a genus of Amazonian siluroid fishes. SIB, vol. 2, no. 3, 

p. 88. November 27. 

1942 

262. Stethaprion innesi and Mylossoma aiireum. T.\, vol. 10, no. 11, pp. 185-186. March. 

[A slightly altered reprint of no. 142.] 

263. [Review of] Longley, W. H., Systematic catalogue of the fishes of Tortugas, Florida, 



36 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

with observations on color, habits, and local distribution. CO, 1942, no. 1, pp. 57-58. 
March 24. [Gives a history of Dr. Longley's submarine ichthylogical researches.] 

264. [Review of] Child, C. M., Patterns and problems of development. CO, 1942, no. 1, 

p. 58. March 24. 

265. The Pacific American atherinid fishes of the genera Eury stole, Nectarges, Coleotropis, 

and Melanorhinus. (G. S. Myers and Charles Barkley Wade.) Allan Hancock 

Pacific Expeditions, vol. 9, no. 5, pp. 113-149, pis. 17-19. March 30. 
*266. A new frog from the Anamallai Hills, with notes on other frogs and some snakes 

from South India. PBSW, vol. 55, pp. 49-55. June 25. 
*267. A new frog of the genus Micrlxalus from Travancore. PBSW, vol. 55, pp. 71-74. 

June 25. 
*268. Notes on Pacific Coast Triturus. CO, 1942, no. 2, pp. 77-82. July 10. 

269. [Review of] Beebe, W., Book of bays. CO, 1942, no. 2, p. 130. July 10. 

269a. [Description of Pimelodella peruana.'] (Carl H. Eigenmann and G. S. Myers.) In: 
Eigenmann, C. H., and Allen, W. R., Fishes of western South America (University 
of Kentucky, Lexington), p. 101, pi. 3. Summer of 1942. 

270. Studies on South-American fresh-water fishes. I. SIB, vol. 2, no. 4, pp. 89-114. 

August 24. 

271. The "lungs" of Bothrioleph. SIB, vol. 2, no. 4, pp. 134-136. August 24. [Virtually 

predicts existence of such an agnathan genus as Jamoytius.] 

*272. The black toad of Deep Springs Valley, Inyo County, California. Occ. Pap. Mus. Zool. 
Univ. Michigan, no. 460, 13 pp., 3 pis. September 16. 

*273. Neotropical lizards in the collection of the Natural History Museum of Stanford Uni- 
versity. (Charles E. Burt and G. S. Myers.) Stanford Univ. Publ., Univ. Ser., Biol. 
Sci., vol. 8, no. 2, pp. 273-324, portrait. October 6. 

*274. Notes on some frogs from Peru and Ecuador. PBSW, vol. 55, pp. 151-155. October 17. 
2 75. [Review of] Marcgrave, J., Historia natural do Brasil. CO, 1942, no. 4, p. 269. 
December 28. [Portugese translation of Marcgrave.] 

1943 

2 76. The Myers Expedition. TA, vol. 11, no. 9, pp. 160-162. January. [Extracts from 

letter from G.S.M. to W. T. Innes on fishes of the Rio Japuhyba, near Angra dos 

Reis, State of Rio de Janeiro, Brazil. Figures are of aquarium fishes, supplied by 

editor.] 

2 77. The influence of Louis Agassiz on the ichthyology of Brazil. Revista Brasileira de 

Biologia, vol. 3, no. 1, pp. 127-133. March. 
278. [Review of] Eigenmann, C. H., and Allen, W. R., Fishes of western South America. 
CO, 1943, no. 1, pp. 60-61. March 31. [Includes comments on the Gregory and 
Conrad classification of Characidae.] 
*279. [Review of] Santos, E., Anfibios e Repteis do Brasil. CO, 1943, no. 1, p. 60. March 31. 
*280. Notes on Rhyacotriton olympicus and Ascaphus truei in Humboldt County, California. 
CO, 1943, no. 2, pp. 125-126. June 30. 
281. George S. Myers reports. TA, vol. 12, no. 6, pp. 104-106. October. [Extracts from 

letter from G.S.M. to W. T. Innes on fishes in Minas Gcrais, Brazil.] 
*282. Rediscovery of the Philippine discoglossid frog, Barboiirula busuangensis. CO, 1943, 

no. 3, pp. 148-150. October 15. 
*283. The lizard names Platyurus and Cosymbotus. CO, 1943, no. 3, p. 192. October 15. 
283a. Sistematica geral de peixes e biologia da pesca. Rio de Janeiro. 84 pp. [This is a 
compiled set of notes taken by several students during a course of lectures given 



Vol. XXXVIII] MYERS: BIBLIOGRAPHY 37 

by G.S.M. at the Museu Xacional in Rio de Janeiro, of which a large number of 
mimeographed copies were issued. The notes were not corrected by G.S.M. before 
mimeographing, and many errors are present.] 

1944 

284. Field notes on fishes of the vicinity of Rio de Janeiro. TA, Philadelphia, vol. 12, no. 

11, pp. 185-186. March. LFigure supplied by editor.] 

285. Field notes on fishes of the vicinity of Rio de Janeiro (concluded). TA, vol. 12, 

no. 12, pp. 204-206. April. [Figure supplied by editor.] 

286. A new species of carangid fish from the northeastern Pacific. (Lionel Albert Walford 

and G. S. Myers.) CO, 1944, no. 1, pp. 44-47. April 21. [Equals Trachiirns 
symmetricus (Ayres), large adults.] 
*287. California records of the western spade-foot toad. CO, 1944, no. 1, p. 58. April 21. 

288. Rhinobrycon negrensis, a new genus and species of characid fishes from the Rio Negro, 

Brazil. Proceedings of the California Academy of Sciences, ser. 4, vol. 2i, no. 39, 
pp. 587-590. August 22. 

289. The black-winged flying characin {Caniegiella marthae Myers). TA, vol. 13, no. 7, 

pp. 105-106. November. [A reprint of 159. See also 186.] 

290. Two extraordinary new blind nematognath fishes from the Rio Negro, representing 

a new subfamily of the Pygidiidae, with a rearrangement of the genera of the family, 
and illustrations of some previously described genera and species from Venezuela 
and Brazil. Proceechngs of the California Academy of Sciences, ser. 4, vol. 23, no. 40, 
pp. 591-602, pis. 52-56. November 7. 
*291. Brazilian books of interest to ichthyologists and herpetologists. CO, 1944, no. 4, 
pp. 262-263. December 26. 

1945 

292. A new gurnard (Prionotus alipionis) from the coast of Brazil. (Gerard Warden 

Teague and G. S. Myers.) Boletim do Museu Nacional, Rio de Janeiro, n. s., zool., 
no. 31, 19 pp. January 24. [Teague wanted to describe this fish although G.S.M. 
was doubtful.] 

293. A remarkable new genus of sexually dimorphic characid fishes from the Rio Paraguay 

Basin in Matto Grosso. (G. S. Myers and Paulo de Miranda-Ribeiro.) Boletim do 
Museu Nacional, Rio de Janeiro, n. s., zool., no. il, 8 pp. January 25. 

*294. Possible introduction of Argentine toads into Florida. CO, 1945, no. 1, p. 44. March 31. 
295. The habitat of Hyphessobrycon flammeiis Myers. FC, vol. 24, no. 10, pp. 73-75. June. 

*296. A natural habitat of the house gecko (Hemidactylus mabouia) in Brazil. CO, 1945, 
no. 2, p. 120. June 30. 

*297. Notes on some new or little-known Brazilian amphibians, with an examination of the 
history of the Plata salamander, Ensatina platensis. (G. S. Myers and Antenor 
Leitao de Carvalho.) Boletim do Museu Nacional, Rio de Janeiro, n. s., zool., 
no. 35, 39 pp. August 25. [Original printed date of issue not correct.] 

*298. A third record of the Sonoran box turtle. CO, 1945, no. 3, p. 172. October 15. 

*299. A strange new leaf-nosed lizard of the genus Anolis from Amazonia. (G. S. Myers and 
Antenor Leitao de Carvalho.) Boletim do Museu Nacional, Rio de Janeiro, n. s., 
zool., no. 43, 22 pp. October 20. 

*300. Nocturnal observations on sea-snakes in Bahia Honda, Panama. Herpetologica, vol. 3, 
no. 1, pp. 22-23. November 2i. 



38 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

1946 

*301. Lista provisoria dos anfibios do Distrito Federal, Brasil. Boletim do Museu Nacional, 
Rio de Janeiro, n. s., zool., no. 55, 36 pp. February 14. [Written in August 1944. 
Text in both Portugese and English.] 

302. The introduction of the guppy (Lebistes) as an aquarium fish, and something on the 

origin of its name. TA, vol. 15, no. 3, pp. 46-48. March. 

303. On a recently proposed new family of deep-sea fishes (Barbourisiidae, Parr, 1945). 

CO, 1946, no. 1, pp. 41-42. April 30. [This and 290 both discuss absence of pelvic 
fins as a taxonomic character.] 

304. Occurrence of uranoscopid fishes of the western Pacific genus Gnathagnus in the 

American Atlantic fauna. CO, 1946, no. 1, p. 42. April 30. 

305. [Review of] Honig, P., and Verdoorn, F., editors, Science and scientists in the 

Netherlands Indies. CO, 1946, no. 1, p. 52. April 30. 

306. New fishes of the families Dactyloscopidae, Microdesmidae, and Antennariidae from 

the west coast of Mexico and the Galapagos Islands with a brief account of the use 
of rotenone fish poisons in ichthyological collecting. (G. S. Myers and Charles 
Barkley Wade.) Allan Hancock Pacific Expeditions, vol. 9, no. 6, pp. 151-179 
(pis. 20-23 included). December 16. 

1947 

307. The Amazon and its fishes. Part 1. The river. AJ, vol. 18, no. 3, pp. 4-9. March 3. 
*308. Murray's Reptiles of Sind, with a note on three forgotten descriptions of Indian sea- 
snakes, published therein. Herpetologica, vol. 3, no. 5, pp. 167-168. [The book 
mentioned is not Murray's Vertebrate Zoology of Sind.] 

309. The Amazon and its fishes. Part 2. The fishes. AJ, vol. 18, no. 4, pp. 13-20. April 1. 

310. [Review of] Smith, H. M., The fresh-water fishes of Siam, or Thailand. CO, 1947, 

no. 1, p. 69. April 20. 

311. [Review of] Hildebrand, S. F., A descriptive catalog of the shore fishes of Peru. CO, 

1947, no. 1, p. 69. April 20. 

312. [Review of] Romer, A. S., Vertebrate paleontology. CO, 1947, no. 1, p. 70. April 20. 

313. [Review of] Woods, R. S., The naturalist's lexicon. CO, 1947, no. 1, pp. 70-71. 

April 20. 

314. The Amazon and its fishes. Part 3. Amazonian aquarium fishes. AJ, vol. 18, no. 5, 

pp. 6-13, 32. May 1. 

315. The varieties of the Siamese fighting fish. AJ, vol. 18, no. 6, pp. 19-21. June 2. 

316. The Amazon and its fishes. Part 4. The fish in its environment. AJ, vol. 18, no. 7, 

pp. 8-19, 34. July. 

317. [Review of] Scale, A., Quest for the golden cloak. CO, 1947, no. 3, p. 213. September 

12. 

318. Foreign introduction of North American fishes — Inadvisability of recommending North 

American fishes without careful appraisal of foreign fishes and ecology. Progressive 
Fish Culturist, Washington, D.C., vol. 9, no. 4, pp. 177-180. October. 

1948 

319. The ramirezi cichlid identified. (G. S. Myers and Robert Rees Harry, Jr.) TA, vol. 

17, no. 4, p. 7. April. [This constitutes the original diagnosis of Apistogramma 
ramirezi; this species probably belongs to Geophagus.] 



Vol. XXXVIII] MYERS: BIBLIOGRAPHY 39 

320. [Review of] Hubbs, C. L., and Lagler, K. F., Fishes of tlie Great Lakes Region. CO, 

1948, no. 2, p. 150. June 30. 

321. Apistogramma ramirezi, a cichlid fish from Venezuela. (G. S. Myers and R. R. Harry, 

Jr.) Proceedings of the California Zoological Club, vol. 1, no. 1, pp. 1-8. August. 
[See 319.] 

322. Notes on two generic names of Indo-Malayan silurid fishes, Wallago and Wallagonia. 

Proceedings of the California Zoological Club, vol. 1, no. 4, pp. 19-20. August. 
[Correction of 212.] 

*323. The California plethodont salamander, Aneides flavipiinctatus (Strauch), with descrip- 
tion of a new subspecies and notes on other western aneides. (G. S. Myers and 
Thomas Paul Maslin.) PBSVV, vol. 61, pp. 127-135. September 3. 

*324. Proposed reprinting of Boulenger's British Museum herpetological catalogues. Herpet- 
ologica, vol. 4, part 5, p. 180. September 13. [See also 325 and 352.] 

*325. Proposed reprinting of Boulenger's British Museum herpetological catalogues. CO, 
1948, no. 3, p. 229. [See also 324 and 352.] 

326. A list of common and scientific names of the better known fishes of the United States 

and Canada. (W. H. Chute, R. M. Bailey, W. A. Clemens, J. R. Dymond, S. F. 
Hildebrand, G. S. Myers and L. P. Schultz.) American Fisheries Society, Special 
Publication no. 1, 45 pp. 

327. [Fundulus in the West Indies.] In: Rivas, L. R., Cyprinodont fishes of the genus 

Fiindidus in the West Indies (Proceedings of the United States National Museum, 
vol. 98, no. 3229, pp. 215-222), pp. 216-217. October 19. [Letter from G.S.M.] 

1949 

328. The Amazon and its fishes. Part 5. A monograph on the piranha [first part]. AJ, 

vol. 20, no. 2, pp. 52-61. February. [See 331.] 
*329. Geographic variation in the ribbed frog, Ascaphits triiei. (M. B. Mittleman and G. S. 

Myers.) PBSW, vol. 62, pp. 57-66. April 27. [Little save introduction is by G.S.M.] 
*330. A new frog of the genus Cormijer from the Solomon Islands, with notes on the 

endemic nature of the Fijian frog fauna. (Walter Creighton Brown and G. S. 

Myers.) AMN, no. 1418, 10 pp. May 9. 

331. The Amazon and its fishes. Part 5. A monograph on the piranha (concluded). AJ, 

vol. 20, no. 3, pp. 76-85. [Published May 31, although issue is for March; see 328.] 

332. Cichlid fishes in salt water. AJ, vol. 20, no. 6, pp. 147-149, 163. June. 

333. Usage of anadromous, catadromous, and allied terms for migratory fishes. CO, 1949, 

no. 2, pp. 89-97. June 30. 

334. Salt-tolerance of fresh-water fish groups in relation to zoogeographical problems. 

Bijdragen tot de Dierkunde, vol. 28, pp. 315-322. August. [In Festschrift for L. L. 
de Beaufort.] 

335. The family name of the characid fishes. CO, 1949, no. 3, pp. 195-204. September 15. 

336. Initial steps in the conservation of fresh-water fisheries in tropical South America, 

with remarks on fishery resources in general. Inter-.'\merican Conference on Con- 
servation of Renewable Natural Resources, held at Denver, Colorado, Sept. 7-20, 
1948, pp. 501-506. October. [Proof not seen by author; references confused by 
editors. Published by the Department of State, Washington, D. C, October 1949.] 
*337. A new frog of the genus Batrachylodes from the Solomon Islands. (Walter Creighton 
Brown and G. S. Myers.) Journal of the Washington Academy of Sciences, vol. 39, 
no. 11, pp. 379-380. November 15. 



40 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

*338. Status of the generic name Lioheterodon applied to Madagascan serpents. Herpetolog- 
ica, vol. 5, part 6, p. 146. December 15. [Text confused by printer; proof not seen 
by author; correctly reprinted as 341.] 

1950 

339. Systematic notes on some Amazonian clupeid fishes of the genus Ilisha. CO, 1950, no. 

1, pp. 63-64. March 30. 

340. On the characid fishes called Hydrocynits and Hydrocyon by Cuvier. Proceedings of 

the CaHfornia Zoological Club, vol. 1, no. 9, pp. 45-47. May 1. 
*341. Status of the generic name Lioheterodon applied to Madagascan serpents. Herpetolog- 

ica, vol. 6, part 2, p. 52. June 5. [No. 338, corrected.] 
*342. Manual of tropical herpetological collecting. 12 pp., mimeographed. Natural History 

Museum, Stanford University. June 15. 

343. Bibliography of the pubhshed papers of George Sprague Myers from 1920 to 1949 

inclusive. 18 pp., mimeographed. June IS. 

344. Identity of the stromateid fish Centrolophus caUfornkiis with Icichthys lockingtoni. 

SIB, vol. 3, no. 4, p. 181. August 21. 

345. Supplementary notes on the flying characid fishes, especially Carnegiella. SIB, vol. 3, 

no. 4, pp. 182-183. August 21. 

346. Studies on South American fresh-water fishes. II. The genera of anostomine characids. 

SIB, vol. 3, no. 4, pp. 184-198. August 21. 

347. A new lump-sucker of the genus Eumicrotremus from the northwestern Atlantic. (G. 

S. Myers and James Erwin Bohlke.) SIB, vol. 3, no. 4, pp. 199-202. August 21. 

348. Station records of the Crocker-Stanford Deep-sea Expedition, Coast of California, 

September 1938. (Rolf Ling Bolin and G. S. Myers.) SIB, vol. 3, no. 4, pp. 203-214. 
August 21. 

349. The "imitator catfish" which mimics a corydoras. (William Thornton Innes and 

G. S. Myers.) TA, vol. 19, no. 9, pp. 222-223. September 1. [See 366.] 
*350. The systematic status of Hyla septentrionalis, the large tree frog of the Florida Keys, 

the Bahamas and Cuba. CO, 1950, no. 3, pp. 203-214. September 5. [See also 447b.] 
*351. [Review of] Bourret, R., Les batraciens de ITndochine. CO, 1950, no. 3, pp. 243-244. 

September 5. 
*352. Proposed reprinting of Boulenger's herpetological catalogues. CO, 1950, no. 3, p. 244. 

September 5. [See also 324 and 325.] 

353. A new genus of poeciliid fishes from Hispaniola, with notes on genera allied to 

Poecilia and Mollienisia. (Luis Rene Rivas and George S. Myers.) CO, 1950, no. 4, 
pp. 288-294. I plate. December 22. 

354. Flying of the halfbeak, euleptorhamphus. CO, 1950, no. 4, p. 320. December 22. 
*355. [Review of] Liu, C. C, Amphibians of western China. CO, 1950, no. 4, p. 325. 

December 22. 
356. [Statement regarding A.S.I.H. Committee on Fish Classification.] CO, 1950, no. 4, 
p. 327. December 22. 

1951 

*357. Notes on salamander voices. CO, 1951, no. 1, p. 76. March 21. 
358. [Review of] Hatch, M. H., editor. Studies honoring Trevor Kincaid. CO, 1951, no. 1, 
pp. 104-105. March 21. 



Vol. XXXVIII] MYERS: BIBLIOGRAPHY 41 

359. [Review of] Tortonese, E., Gli animali superiori nella loro struttura e nella loro vita. 

CO, 1951, no. 1, p. 106. March 21. 

360. Study of fishes was the work David Starr Jordan loved best. Stanford Alumni Re- 

view, vol. 52, no. 7, pp. 13-15. March 27. 

361. The Amazonian mottled knife-fish, Steatogenys elegans, and its strange vermiform 

organ. TA, vol. 20, no. 4, pp. 85-86. April 10. 

362. The Amazonian checkerboard cichlid (Crenicara maculata) . TA, vol. 20, no. 5, pp. 

109-110. May 1. [See 367.] 

363. Notas sobre la distribucion de los peces Sudamericanos del grupo Bivibranchia. 

Memorias Sociedad de Ciencias Xaturales La Salle, Caracas, tomo 10, no. 27 (for 

Sept.-Dec. 1950), pp. 193-194. [Published in late Spring, 1951.] 
*364. The most widely heard amphibian voice. CO, 1951, no. 2, p. 179. June 8. [Voice of 

Hyla regilla in Hollywood sound-cinema.] 
*365. Asiatic giant salamander caught in the Sacramento River, and an exotic skink near San 

Francisco. CO, 1951, no. 2, pp. 179-180. June 8. 

366. The "imitator catfish" which mimics a corydoras. (Wilham Thornton Innes and G. S. 

Myers.) In: Innes, W. T. (editor). Aquarium Highlights, consisting of reprints of 
the most popular articles from the monthly magazine, the Aquarium, since 1932 
(Innes Publishing Co., Philadelphia, 519 pp.), pp. 117-118. October. [Reprint of 
349.] 

367. Amazon dwarf checkerboard cichhd. In: Innes, W. T. (editor). Aquarium Highhghts 

(see 366 for full reference), pp. 120-121. October. [Reprint of 362.] 

368. Dangers in identifications. In: Innes, W. T. (editor). Aquarium Highlights (see 366 

for full reference), pp. 366-371. October. [Reprint of 139.] 

369. Judging fish shows. In: Innes, W. T. (editor). Aquarium Highhghts (see 366 for full 

reference), pp. 410-413. October. [Reprint of 169.] 

370. David Starr Jordan, ichthyologist, 1851-1931. SIB, vol. 4, no. 1, pp. 2-6. December 27. 

371. Fresh-water fishes and East Indian zoogeography. SIB, vol. 4, no. 1, pp. 11-21. De- 

cember 27. [See 483.] 

372. Some forgotten but available names for Indian fishes. SIB, vol. 4, no. 1, p. 26. De- 

cember 27. 
*373. A new giant toad from Southwestern Colombia. (G. S. Myers and John W. Funk- 
houser.) Zoologica, New York, vol. 36, pt. 4, pp. 279-281, 1 pi. December 28. 
[Bufo blombei'gi, n. sp.] 

1952 

374. Tower of Babel. [Editorial.] CO, 1952, no. 1, pp. 57-58. June 2. 

375. [Obituary of] William G. Holbein. AJ, vol. 23, no. 6, pp. 118-119. June 12. 

376. Danio or Brachydanio, Barbus or Puntius? AJ, vol. 23, no. 6, pp. 121-122. June 12. 

[Vol. number erroneously given as 24 on the issue.] 

377. [Review of] Axelrod, H. R., Tropical fish as a hobby. CO, 1952, no. 2, pp. 120-121. 

June 26. 

378. [Review of] Ladiges, W., Der Fisch in der Landschaft. CO, 1952, no. 2, p. 121. June 

26. 

379. [Review of] Steward, J. H., Handbook of South American Indians. Vol. 6. CO, 1952, 

no. 2, pp. 121-122. June 26. 

380. [Review of] Carter, G. S., Animal evolution, a study of recent views on its causes. 

CO, 1952, no. 2, p. 122. June 26. 



42 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

381. [Review of] Kuenen, P. K., Marine geology. CO, 19S2, no. 2, p. 122. June 26. 

382. [Review ofj Suwatti, C, Fauna of Thailand. CO, 1952, no. 2, pp. 122-123. June 26. 

383. [Review of] Primer Congresso Nacional de Pesquerias Maritimas y Industrias Deri- 

vadas. CO, 1952, no. 2, p. 123. June 26. 

384. [Review of] Soljan, T., Fauna et Flora Adriatica, vol. 1, Pisces. CO, 1952, no. 2, p. 

123. June 26. 
*385. [Review of] Stebbins, R. C, Amphibians of Western North America. CO, 1952, no. 2, 

pp. 123-124. June 26. 
386. [Obituary of] Chloe Leslie Starks. CO, 1952, no. 2, pp. 124-125. June 26. [America's 

most able fish illustrator.] 
*387. A new dwarf toad from southeastern Brazil. (G. S. Myers and Antenor Leitao de 

Carvalho.) Zoologica, New York, vol. 37, no. 1, pp. 1-3. June 30. \Bufo pygmaejis, 

n. sp.] 

388. Annual fishes. AJ, vol. 2i, no. 7, pp. 125-141. July 11. 

389. [Obituary of] Louis L. Mowbray. AJ, vol. 2i, no. 7, p. 141. July 11. 

389a. On the problem of the status of names published by Scopoli in 1777 in his "Introduc- 
tio ad Historiam Naturalem." Bulletin of Zoological Nomenclature, vol. 6, pt. 8, p. 
255. July 23. [Reprinted in: Opinions and Declarations, vol. 9, pt. 23, p. 316.] 

390. BibUography of the published papers of George Sprague Myers from 1920 to 1951, in- 

clusive. 20 pp., mimeographed. [Dated May 1951; issued during July 1952.] 

391. [Review of] Gohm, D., Tropical fish in the home. /\J, vol. 23, no. 8, pp. 152-153. 

August 7. 
*392. [Review of] Carr, A., Handbook of turtles. AJ, vol. 25, no. 8, p. 153. August 7. 

393. Hints to fish importers, no. 1. AJ, vol. 23, no. 8, pp. 156-157. August 7. 

394. [Obituary of] Johann Paul Arnold, 1868-1952. AJ, vol. 23, no. 9, pp. 169-170. Au- 

gust 20. 

395. Hints to fish importers, no. 2. AJ, vol. 23, no. 9, pp. 171-173. August 28. [Lake 

Tanganyika; helped initiate the reports of Tanganyikan aquarium fishes.] 

396. A note on the feathertail, an African characin {Phenacogrammm) . AJ, vol. 23, no. 9, 

p. 173. August 28. 

397. [Review of] Hervey, G. F., and Hems, J., Freshwater tropical aquarium fishes. AJ, 

voL 23, no. 9, pp. 174-177. August 23. 

398. The nature of systematic biology and of a species description. Systematic Zoology, vol. 

1, pp. 106-111. September. 

399. Easiest of all to spawn and raise, the medaka. AJ, vol. 23, no. 10, pp. 189-194. Sep- 

tember 30. [Care, spawning, and raising Oryzias latipcs.] 

400. Varieties of the three-spot gourami, Trichogaster trichopterus. AJ, vol. 23, no. 10, pp 

198-200. September 30. 

401. [Review of] Beck, P., Traite complet de la vie des animaux en aquarium. A J, vol. 23 

no. 10, p. 200. September 30. 

402. Amazonian tetras of the genus Thayeria. AJ, vol. 23, no. 10, pp. 206-207. September 

30. [Species here called T. sanctae-mariae later described as T. boehlkei Wcitzman 
See also 429.] 

403. How the shooting apparatus of the archer fish was discovered. AJ, vol. 23, no. 10, pp 

210-214. September 30. 

404. Hints to fish importers, no. 3. AJ, vol. 23, no. 10, pp. 215-216. September 30. [Bur- 

mese fishes.] 

405. A new Amazonian catfish for the aquarist. AJ, vol. 23, no. 11, pp. 224-225. October 

31. [Corydoras elegans.] 



Vol. XXXVIII] MYERS: BIBLIOGRAPHY 43 

406. Color schemes in your fishes. AJ, vol. 23, no. 11, pp. 228-230. October 31. 

407. Hints to fish importers, no. 4. AJ, vol. 23, no. 11, pp. 237-238. October 31. [His- 

paniola.] 

408. [Review of] Ladiges, W., Der Fisch in der Landschaft. AJ, vol. 23, no. 11, p. 239. 

October 31. 

409. [Review of] Tropical Fish Hobbyist. AJ, vol. 25, no. 11, pp. 239-240. October 31. 

410. Corydoras elegans. TA, Philadelphia, vol. 21, no. 11, pp. 300-301. November. 

411. [Re\iew of] L'Aquarium Exotique. AJ, vol. 23, no. 12, pp. 254-255. December 1. 

412. [Review of] Wendt, A., Die Aquarienpflanzen in Wort und Bild. AJ, vol. 23, no. 12, 

p. 255. December 1. 

413. [Review of] Holly, M., Meinken, H., and Rachow, A., Die Aquarienfische in Wort und 

Bild. AJ, vol. 23, no. 12, p. 255. December 1. 

414. [Review of] Lorenz, K., King Solomon's ring. AJ, vol. 23, no. 12, p. 256. December 1. 

415. Hints to fish importers, no. 5. AJ, vol. 23, no. 12, p. 256. December 1. [Fitudnlus 

stelUfer.] 

416. [Obituary of] Dr. Ernst Bade. AJ, vol. 23, no. 12, p. 264. December 1. 

417. [Review of] Innes, W. T., Exotic aquarium fishes, 15th edition. AJ, vol. 24, no. 1, p. 

9. December 19. 

418. The miniature fish aquarium. AJ, vol. 24, no. 1, pp. 11-12. December 19. 

419. Hints to fish importers, no. 6. AJ, vol. 24, no. 1, p. 20. December 19. [Sierra Leone; 

Epiplatys anuulatus.] 

420. Sharks and sawfishes in the Amazon. CO, 1952, no. 4, pp. 268-269. December 26. 

421. [Review of] Deraniyagala, P. E. P., A colored atlas of some vertebrates from Ceylon. 

CO, 1952, no. 4, p. 286. December 26. 

1953 

422. Spawning behavior of Polycentnis. AJ, vol. 24, no. 2, pp. 31-33. January 28. 

423. Hints to fish importers, no. 7. AJ, vol. 24, no. 2, pp. 33-34. January 28. [Chologaster.] 
*424. [Review of] Knight, M., Keeping reptiles and fishes. A J, vol. 24, no. 2, p. 46. January 

28. 

425. [Review of] Whitney, L. F., All about guppies. AJ, vol. 24, no. 3, p. 55. February. 

426. [Aquarium water-testers, filters, oil traps, and thermostatic heaters.] AJ, vol. 24, no. 

3, pp. 60-61. February 23. 

427. Hints to fish importers, no. 8. Two beautiful httle characins (Nematobrycon) from 

western Colombia and their peculiar distribution. AJ, vol. 24, no. 3, pp. 64-65. Feb- 
ruary 23. [This initiated importation of the "Emperor tetra" as an aquarium fish.] 

428. The living-fossil coelacanth fishes. AJ, vol. 24, no. 3, pp. 66-68. February 23. 

429. Vinkeltetrornas vetenskapliga nanm. Akvariet (Organ for Sveriges Akvarieforeningar) , 

Argang 27, no. 3, pp. 42-44. March. [See also 402.] 

430. Queens of the water. AJ, vol. 24, no. 4, pp. 75-78. March 30. [An essay on water- 

Ulies: Nuphar, Nelnmbo, Nymphaea, Victoria, and Euryale.] 

431. Pets. [Verse.] AJ, vol. 24, p. 78. March 30. 

432. Hints to fish importers, no. 9. AJ, vol. 24, no. 5, p. 112. April 29. [Moenkhausia 

costae.] 

433. [Review of] Whitney, L. F., The complete book of home pet care. AJ, vol. 24, no. 5, 

p. 119. April 29. 

434. [Review of] Knowles, F. G. W., Freshwater and saltwater aquaria. AJ, vol. 24, no. 5, 

p. 119. April 29. 



44 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

435. [Review of] Ladiges, W., Zierfisch Bilderbuch. AJ, vol. 24, no. 5, p. 119. April 29. 

436. [Review of] Kramer, K., and Weise, H., Aquarienkunde. AJ, vol. 24, no. 5, pp. 119- 

120. April 29. 

437. The coelacanth fishes — living fossils. TA, vol. 22, no. 5, pp. 145-146. May. [Figure 

erroneously labeled.] 

438. Aquarium difficulties with black mollienesias. AJ, vol. 24, no. 6, pp. 125-129. May 29. 

[This paper was later reprinted at least twice in the same journal, but references are 
not at hand.] 

439. Hints to fish importers, no. 10. A strange glandulocaudine characin from the Rio das 

Velhas (Hysteronotus) . AJ, vol. 24, no. 6, p. 137. May 29. 
*440. [Review of] Leutscher, A., Vivarium hfe. AJ, vol. 24, no. 6, p. 141-142. May 29. 

441. [Review of] Evans, A., Aquariums. AJ, vol. 24, no. 6, p. 142. May 29. 

442. [Review of] Ichthys. AJ, vol. 24, no. 7, p. 157. June 29. 

443. [Review of] Tropical fish handbook-catalog. AJ, vol. 24, no. 7, pp. 157-158. June 29. 

444. [Review of] Taschenkalender fiir Aquarien und Terrarienfreunde, 1953. AJ, vol. 24, 

no. 7, p. 158. June 29. 

445. [Review of] Rounsefell, G. A., and Everhart, W. H., Fishery science; its methods and 

applications. AJ, vol. 24, no. 7, p. 158. June 29. 

446. [Review of] New Zealand Aquatic World. AJ, vol. 24, no. 7, p. 158. June 29. 

447. Hints to fish importers, no. 11. Glass catfishes. AJ, vol. 24, no. 7, pp. 161-162. June 

29. [Kryptoptencs, Ailiichthys, Parailia, Physailia, Psetidepapteriis. Evolutionary 
convergence in 3 families of catfishes.] 
447a. [Quotation from letter regarding retention of original spellings of zoological names.] 

Bulletin of Zoological Nomenclature, vol. 10, pt. 7, p. 216. July 14. 
*447b. [On the acceptance of certain names originally published in synonymy, especially 
Eleutherodactylus and Hyla septentrionalis.] Bulletin of Zoological Nomenclature, 
vol. 10, pts. 10-11, p. 312. July 24. 

448. Piabucus in the aquarium. AJ, vol. 24, no. 8, pp. 172-174. July 28. [Identification of 

live examples of Piabucus and similar genera.] 

449. Unhanded color variety of the Malayan "coolie" loach, Acanthophthalmus semkinctus. 

AJ, vol. 24, no. 8, p. 174. July 28. 

450. The Florida pigmy topminnow, Leptolucania ominata, its history, and a record of the 

first California breeding. AJ, vol. 24, no. 8, pp. 184-187. July 28. 

451. On aquarium magazines. [Editorial.] AJ, vol. 24, no. 8, pp. 188-189. July 29. 

452. The Cuban green glass fish, Atherina evermanni. AJ, vol. 24, no. 9, pp. 195-197. Sep- 

tember. [For correction of name to Alepidomus evermanni, see same journal, vol. 
24, no. 11, p. 170.] 

453. Why show standards for most tropical fishes are unwise. AJ, vol. 24, no. 9, pp. 200- 

202. September. 

454. Publication dates of the Aquarium Journal. | In 1952-1953.] AJ, vol. 24, no. 9, p. 202. 

September. 

455. Hints to fish importers, no. 12. Rivuliis ornatus. AJ, vol. 24, no. 9, p. 208. September. 

456. Neon tetra. A J, vol. 24, no. 9, pp. 210-211. September. 

457. [Obituary of] Floyd S. Young. AJ, vol. 24, no. 9, pp. 211-212. September. 

458. What's wrong with aquarists? Tropical Fish Tales, Springfield, Mass., vol. 2, no. 4, 

pp. 6-7. September. 

459. Zebra danio. A J, vol. 24, no. 10, pp. 230-231. October. \Brachydamo rerioA 



Vol. XXXVIII] MYERS: BIBLIOGRAPHY 45 

460. Classification of the danios. AJ, vol. 24, no. 10, pp. 235-238. October. [Critical taxo- 

nomic evaluation of four genera, Danio, Daniops, Allodanio, Brachydanio. See also 
513a.] 

461. Where, oh where do the pictures come from? [Editorial.] AJ, vol. 24, no. 10, pp. 

242-243. October. 

462. Hints to fish importers, no. 13. Rivulus zygonectes. .\J, vol. 24, no. 10, p. 244. October. 

463. Notes on selecting an aquarium. AJ, vol. 24, no. 11, pp. 254-256. November. 

464. [Review of] Innes, W. T., Exotic aquarium fishes, 16th Edition. AJ, vol. 24, no. 11, 

p. 258. November. 

465. Serpa tetra. AJ, vol. 24, no. 11, pp. 264-265. November. [Hyphessobrycon callistus.] 

466. Hints to fish importers, no. 14. Garmanella pulchra. AJ, vol. 24, no. 11, p. 266. No- 

vember. 

467. A note on the habits and classification of Corydoras hastatiis. AJ, vol. 24, no. 11, pp. 

268-270. November. [Subgenus Mkrocorydoras.'] 

468. The Christmas-tree fish. [Fantasy.] AJ, vol. 24, no. 12, p. 280. December. 

469. The secrets of the German fish breeders. [Editorial.] AJ, vol. 24, no. 12, pp. 287-288. 

December. 

470. Hints to fish importers, no. 15. Barbus candens. AJ, vol. 24, no. 12, p. 296. December. 

471. Habits of the spotted knife-fish (Steatogenys elegans) in the aquarium. AJ, vol. 24, 

no. 12, pp. 297-298. December. 

1954 

472. Hints to fish importers, no. 16. A fish that hasn't been discovered yet. AJ, vol. 25, no. 

1, p. 5. January. [Predicts the occurrence of a relative of Cubanichthys and Chrio- 
peopoides in Hispaniola.] 

473. [Review of] Todd, R., The tropical fish book. AJ, vol. 25, no. 1, p. 7. January. 

474. Siamese fighting fish. AJ, vol. 25, no. 2, pp. 27-29. February. 

475. The fighting fish and its history. AJ, vol. 25, no. 2, pp. 30-33. February. [Betta 

splendens.] 

476. Hints to fish importers, no. 17. AJ, vol. 25, no. 2, p. 39. February. [Wild stock of 

Betta splendens.] 

477. [Review of] Emmens, C. W., Keeping and breeding aquarium fishes. AJ, vol. 25, no. 

2, pp. 45—46. February. 

478. Fifty years of devotion to the aquarium hobby. TA, vol. 2i, no. 2, pp. 35-39. Febru- 

ary. [Tribute to WilHam Thornton Innes, on his 80th birthday.] 

479. Protective coloration in the leaf fish and Thayeria. AJ, vol. 25, no. 3, pp. 62-63. March. 

480. [Review of] Roberts, J. B., Jr., The pet shop manual. AJ, vol. 25, no. 3, pp. 77-78. 

March. 

481. [Review of] The Tropical Fish Magazine. AJ, vol. 25, no. 3, p. 78. March. 

*482. AbiUty of amphibians to cross sea barriers, with especial reference to Pacific zoogeog- 
raphy. Proceedings 7th Pacific Science Congress [New Zealand, February 1949], vol. 
4 (zoology), pp. 19-27. [Dated 1953; published March 1954.] 

483. Paleogeographical significance of fresh-water fish distribution in the Pacific. Proceed- 

ings 7th Pacific Science Congress [New Zealand, February 1949], vol. 4 (zoolog\). 
pp. 38-48. [When printing of the Proceedings had already been delayed for nearly 
three years, the present paper was printed in the United States as 371. Vol. 4 of the 
Proceedings finally was pubhshed (dated 1953) in March 1954.] 

484. The protection of rare and vanishing fishes. Proceedings 7th Pacific Science Congress 



46 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

[New Zealand, February 1949], vol. 4 (zoology), pp. 691-694. [Dated 1953; pub- 
lished March 1954. Aside from the Austrahan lungfish, this represents the first plea 
for the conservation of non-food, non-game fishes.] 

485. Blue gularis [Aphyosemion caeruleutn.] AJ, vol. 25, no. 4, pp. 87-88. April. 

486. How to preserve fish specimens for study. AJ, vol. 25, no. 4, pp. 89-90. April. 

487. Another new corydoras from Brazil. (G. S. Myers and Stanley Howard Weitzman.) 

AJ, vol. 25, no. 4, pp. 93-94. April. [Corydoras cochui, n. sp.] 

488. Hints to fish importers, no. 18. AJ, vol. 2S, no. 4, p. 102. April. [Poecilocharax 

bovallii.] 

489. A new corydoras. (G. S. Myers and William Thornton Innes.) TA, vol. 23, no. 4, p. 

105. April. [Corydoras cochui Myers and Weitzman. See 487 for original descrip- 
tion.] 
489a. [Supplementary note.] Opinions and Declarations rendered by the International Com- 
mission on Zoological Nomenclature, vol. 4, pt. 15, p. 167. April 21. [Concerns 
Raphistoma versus Belone.] 

490. Hints to fish importers, no. 19. An unknown characin from the Cerro Duida, Vene- 

zuela. AJ, vol. 25, no. 5, pp. 111-112. May. [A lost and still undescribed relative of 
Poecilocharax.^ 

491. The Amazon longfin, Pterolebias longipinnis. (Fritz Mayer and G. S. Myers.) AJ, vol. 

25, no. 5, pp. 113-115. May. 

492. [Review of] Fisher, E. L., Marine tropicals. AJ, vol. 25, no. 5, pp. 126-128. May. 

493. [Review of] La vita nell 'acquario. Manuale catalogo dell 'Acquario di Bologna. AJ, 

vol. 25, no. 5, p. 128. May. 

494. The black-banded sunfish. A J, vol. 25, no. 6, pp. 133-134. June. [Mesogomstius 

chaetodon.] 

495. The name of the Indian glassfish (Chanda lala) . AJ, vol. 25, no. 6, pp. 149-150. June. 

496. The kissing gourami. AJ, vol. 25, no. 7, pp. 155-156. July. [Helostoma temminckii.] 

497. A new cyprinodont fish from the Peruvian Amazon. AJ, vol. 25, no. 8, pp. 175-177. 

July 28. [Pterolebias peruensis, n. sp.] 

498. [Review of] Nachstedt, J., and Tusche, H., Breeding aquarium fishes. AJ, vol. 25, no. 

8, p. 183. July 28. 

499. A beautiful new cyprinodont fish from the Amazon. TA, vol. 23, no. 8, pp. 236-237. 

August. [Pterolebias peruensis; see 497 for original description.] 

500. The life and times of Polycentrus and the leaf-fish tribe. Tropical Fish Magazine, 

Springfield, Mass., vol. 4, no. 2, pp. 6-7. October. [Suggests possible relationship 
between Nandidae, Datnioides, and Lobotes.] 

501. What fish is that? Tropical Fish Magazine, Springfield, Mass., vol. 4, no. 3, pp. 8-9. 

November. 

1955 

501a. Notes on the freshwater fish fauna of middle Central America, with especial reference 
to pond culture of Tilapia. Fish Papers, FAO, Rome, no. 2. 1955. [Pagination and 
exact date lacking; warns of dangers in Tilapia introductions.] 

501b. [Limericks] In: Martin, H. R. (editor). The little Hmerick book (Peter Pauper Press, 
Mt. Vernon, N. Y., 62 pp.). 1955. 

502. Notes on the classification and names of cyprinodont fishes. Tropical Fish Magazine, 

Springfield, Mass., vol. 4, no. 7, p. 7. March 1. [Includes diagnoses of Pantanodon 
podoxys Myers, n. g., n. sp., and Polamophylax pygmaeus Myers and Carvalho, n. 
gen., n. sp. Proof not seen by author. There are a few editorial and printer's errors.] 



Vol. XXXVIII] MYERS: BIBLIOGRAPHY 47 

503. English names of aquarium cichlids. Tropical Fish Magazine, Springfield, Mass., vol. 

4, no. 7, pp. 8-9. March 1. [On p. 9, two columns of print are transposed.] 

504. The wonderful world under the sea. Part 1. Tropical Fish Magazine, Springfield, Mass., 

vol. 4, no. 8, pp. 8-9. April. 

505. Gambusinos — a new term proposed for "live-bearing toothcarps." TA, vol. 24, no. 5, 

pp. 149-152. May. 

506. The wonderful world under the sea. Part 2. Tropical Fish Magazine, Springfield, 

Mass., vol. 4, no. 9, pp. 11, 14. May. 

1956 

*S07. Manual of tropical herpetological collecting. Ed. 2. Natural History Museum of Stan- 
ford University, Circular no. 4, 13 pp. [Mimeographed; very limited edition; second 
edition of 342.] 

*S08. Brief directions for preserving and shipping specimens of fishes, amphibians and rep- 
tiles. Natural History Museum of Stanford University, Circular no. 5, 3 pp. [Mime- 
ographed; very limited edition.] 

509. [Comments on Axelrod, H., and Schultz, L. P., Handbook of tropical fishes.] TA, vol. 

25, no. 2, pp. 60-61. February. [Quotes from letter from G. S. Myers.] 

510. Two new BraziUan fresh-water fishes. (G. S. Myers and Stanley Howard Weitzman.) 

SIB, vol. 7, no. 1, pp. 1-4. February 21. [Hyphessobrycon cardinalis and Hassar 
praelongus. H. cardinalis was also described by Schultz as Cheirodon axelrodi, in a 
pubhcation bearing the printed date February 20.] 

511. The name of the South American clupeid fish. Prist igaster. CO, 1956, no. 1, pp. 63-64. 

February 29. 
*S12. [Review of] Cochran, D. M., Frogs of southeastern Brazil. CO, 1956, no. 1, p. 69. 

February 29. 
*S13. Zoological results of the California Himalayan Expedition to Makalu, Eastern Nepal. 

I. Amphibians and Reptiles. (Alan E. Leviton, G. S. Myers, and Lawrence W. 

Swan.) Occasional Papers of the Natural History Museum of Stanford University, 

no. 1, 18 pp. March 9. 
513a. Classification des danios. L'Aquarium et les Poissons, Paris, 6me Annee, no. 4, pp. 5- 

8. April. [Translation of 460.1 

514. Note on guppies in Mexico. TA, vol. 25, no. 4, p. 123. .^pril 

515. Studies on the fishes of the family Characidae. No. 11. A new genus and species of 

hemiodontins from the Rio Orinoco in Venezuela. (James Erwin Bohlke and G. S. 
Myers.) Notulae Naturae, Philadelphia, no. 286, 6 pp. May 23. 

516. Request for a ruUng as to the species to be accepted as the type species of the genera 

Culter and Nasus Basilewsky, 1855 (class Pisces). Bulletin of Zoological Nomencla- 
ture, vol. 12, pp. 136-138. July. [Francis Hemming, Secretary of the I.C.Z.N., re- 
worded 243, added certain requests to it, and published it under the name of G. S. 
Myers. Reprinted in: Opinions and Declarations, 1958, vol. 18, pt. 17, pp. 294-297.] 

517. The Xenurobryconini, a group of minute South American characid fishes with teeth 

outside the mouth. (G. S. Myers and James Erwin Bohlke.) SIB, vol. 7, no. 2, pp. 
6-12. August 30. 

518. Copella, a new genus of pyrrhulinin characid fishes. SIB, vol. 7, no. 2, pp. 12-13. Au- 

gust 30. 

519. Esomtis rehi, an Indo-Malayan cyprinid fish. SIB, vol. 7, no. 2, pp. 13-14. August 30. 

[Pogonocharax rehi Regan, described as from "Argentina."] 



48 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

520. A note on an Abyssinian catfish, Clarias depressus Myers. SIB, vol. 7, no. 2, p. 14. 

August 30. [See 45.] 

521. Curatorial practices in zoological research collections. 2. System followed in filing 

specimens of Recent fishes in the Natural History Museum of Stanford University. 
(G. S. Myers and Margaret Hamilton Storey.) Natural History Museum of Stan- 
ford University, Circular no. 6, 44 pp. October. [Mimeographed; very small edi- 
tion.] 

1957 

522. Exotic aquarium fishes — a work of general reference. By William Thornton Innes. 

19th Edition, revised, enlarged, and edited by G. S. Myers. Innes Publishing Com- 
pany, Philadelphia; 541 pp., colored frontispiece, 90 colored figs., 366 black and 
white figs., 7 maps, 2 maps on linings. [All previous 18 editions of this book were 
also revised by G.S.M. before publication, but only in this edition was editorial re- 
sponsibility formally assumed. Previous editions are not listed in this bibHography.] 

1958 

523. [Four world maps, showing extent of reef coral, current knowledge of species occur- 

rence of marine fishes, current knowledge of habits of marine food fishes, and the 
species composition of marine food-fish faunas.] In: Walford, L. A., Living re- 
sources of the sea — opportunities for research and expansion (New York, Ronald 
Press Co., xvi -f 231 pp.), figs. 16-19, with discussions on pp. 246, 248, and 250. 
[Original maps and discussions by G.S.M. Other research for this book was also 
done by G.S.M.] 

524. Trends in the evolution of teleostean fishes. SIB, vol. 7, no. 3, pp. 27-30. July 31. 

[Paper presented before Society for the Study of Evolution, August 1957.] 

525. Nomenclator of certain terms used for higher categories of fishes. SIB, vol. 7, no. 3, 

pp. 31-40. July 31. 

526. The priacanthid fish genus Pristigenys. SIB, vol. 7, no. 3, pp. 40-42. July 31. 

1959 

527. The endemic fish-fauna of Lake Lanao and the evolution of higher categories. Pro- 

ceedings XVth International Congress of Zoology (London, 1958), pp. 151-152. [Ab- 
stract of 537.] 

528. A remarkable new genus of anostomin characid fishes from the upper Rio Xingi'i in 

central Brazil. (G. S. Myers and Antenor Leitao de Carvalho.) CO, 1959, no. 2, 
pp. 148-152. July 24. \Sartor rcspectus, n. gen., n. sp.] 

529. A Caribbean chaetodont fish, Chaetodon eqiies Steindachner, now referred to Chaeto- 

don aya Jordan. CO, 1959, no. 2, p. 158. July 24. 

530. [Review of] Simpson, G. G., and Roe, A., Behavior and evolution. CO, 1959, no. 3, 

pp. 270-271. October 9. 

1960 

531. Restriction of the croakers (Sciaenidae) and anchovies (Engraulidae) to continental 

waters. CO, 1960, no. 1, pp. 67-68, March 25. 



Vol. XXXVIII] MYERS: BIBLIOGRAPHY 49 

*S32. Phylax telescopus. CO, 1960, no. 1, pp. 75-78. March 25. [A column of comment, 
criticism, and review dealing not only with ichthyology and herpetology but also 
broader subjects.! 

533. [Review of] Whitley, P. G., and .Mien, J., The sea-horse and its relatives. CO, 1960, 

p. 78. March 25. 

534. I Review of] Poll, M., Les genres des poissons d'eau douce de I'Afrique. CO, 1960, 

no. 1, pp. 78-79. March 25. 
*S35. Phylax telescopus, II. CO, 1960, no. 2, pp. 157-159. June 29. [Darwin; ideas versus 
data; the eel problem; Ahpio de Miranda-Ribeiro, etc. In footnote, for Freund read 
Freud.] 

536. [Review of] Mclnerny, D., and Girard, G., All about tropical fish. CO, 1960, no. 2, 

p. 162. June 29. 

537. The endemic fish fauna of Lake Lanao, and the evolution of higher taxonomic cate- 

gories. Evolution, vol. 14, no. 3, pp. 323-333. September. [Proof not seen by au- 
thor; p. 328, paragraph 2, for biotypes read biotopes; p. 330, delete entire sentence 
containing word "Triglopsis." See also 527 for abstract; also 591 for reprintings.] 

538. Fish evolution in Lake Nyasa. Evolution, vol. 14, no. 3, pp. 394-396. September. [Di- 

versity of freshwater fish faunas of the world, etc.] 

539. Some reflections on phylogenetic and typological taxonomy. Systematic Zoology, vol. 

9, no. 1, pp. 37-41. [March 1960; published September I960.] 
*540. Phylax telescopus. III. CO, 1960, no. 3, pp. 263-266. September 26. [On borrowing 
specimens; electric fishes; frog phylogeny, etc.] 

541. [Review of] Poll, M., Expedition oceanographique Beige . . . I'Atlantic sud. Poissons. 

CO, 1960, no. 3, pp. 267-268. September 26. 

542. [Review of] Parr, A. E., Mostly about museums. CO, 1960, no. 3, p. 268. September 

26. 

543. [Obituary of] Margaret Hamilton Storey (1900-1960). SIB, vol. 7, no. 4, p. 62a. Oc- 

tober 27. [See 557 for another obituary of M.H.S.] 

544. A new zeomorph fish of the family Oreosomatidae from the coast of California, with 

notes on the family. SIB, vol. 7, no. 4, pp. 89-98. October 27. 

545. Two new fishes collected by General Thomas D. White in eastern Colombia. (G. S. 

Myers and Stanley Howard Weitzman.) SIB, vol. 7, no. 4, pp. 98-109. October 27. 

546. The mormyrid genera Hippopotamyrus and Cyphomyrus. SIB, vol. 7, no. 4, pp. 123- 

125. October 27. 

547. The genera and ecological geography of the South American banjo catfishes, family 

Aspredinidae. SIB, vol. 7, no. 4, pp. 132-139. October 27. 

548. A Brazilian pike-characid, Boulengerella lateristriga, rediscovered in the Rio Negro. 

(G. S. Myers and Stanley Howard Weitzman.) SIB, vol. 7, no. 4, pp. 201-205. Oc- 
tober 27. 

549. The South American characid genera Exodon, Gnathoplax, and Roeboexodon, with 

notes on the ecology and taxonomy of characid fishes. SIB, vol. 7, no. 4, pp. 206- 
211. October 27. 

550. Preface to any future classification of the cyprinid fishes of the genus Barbus. SIB, 

vol. 7, no. 4, pp. 212-215. October 27. 

551. A forgotten account of a fresh-water belonid fish from northern India. SIB, vol. 7, no. 

4, pp. 345-346. October 27. 

552. The names of the South American catfish genera Conorhynchos and Diplomystes. SIB, 

vol. 7, no. 4, pp. 246-248. October 27. [See errata on p. 62b of same issue. Cono- 
rhynchos misspelled in three places on p. 247.1 



50 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

*SS3. Phylax telescopus, IV. CO, 1960, no. 4, pp. 373-377. December 30. [Oceanography 
and the neglect of biological collecting; European centers; Denticeps, etc.] 

1961 

*554. Phylax telescopus, V. CO, 1961, no. 1, pp. 117-120. March 17. [Conservationists 
neglect fishes; a new living perch from Europe; etc.] 

*555. The South American hyhd frog names Sphaenorhynchns, Dryomelictes, and Spho- 
enohyla. (G. S. Myers and Alan Edward Leviton.) Herpetologica, vol. 17, pp. 61-62. 
April 15. 

*5S6. Phylax telescopus, VI. CO, 1961, no. 2, pp. 244-247. June 19. [Fishery biology and 
management; European museums, etc. Proof not seen by author; for Wandsee read 
Wansee.] 
557. [Obituary of] Margaret Hamilton Storey (1900-1960). CO, 1961, no. 2, pp. 261-263. 
June 19. [Not the same as 543.] 

*S58. Generic type species citation in taxonomic zoology. A guide for students. Natural 
History Museum of Stanford University, Circular no. 8, 7 pp. August. [Mimeo- 
graphed, very small edition. Also reproduced later by U.S. Bureau of Commercial 
Fisheries Ichthyological Laboratory, U.S. National Museum, Washington, D.C.] 

*559. The New Zealand lizard names Nauliinus and Hoplodactylm. Herpetologica, vol. 17, 
no. 3, pp. 169-172. October 9. 

1962 

*S60. The American leptodactylid frog genera Eleutherodactylus, Hylodes (-Elosia), and 
Caiidiverbera {= Calyptocephalus) . CO, 1962, no. 1, pp. 195-202. April 11. [Suako 
447b.] 
561. Statement regarding the argument of W. I. Follett and Daniel M. Cohen concerning 
the type species of the genus Bathylagiis. Bulletin of Zoological Nomenclature, vol. 
19, pp. 130-131. May 28. [Deahng principally with the type designations of Jordan 
and Evermann.] 

*562. The Hong Kong newt described as a new species. (G. S. Myers and Alan Edward 
Leviton.) Occasional Papers, Division of Systematic Biology [formerly Natural His- 
tory Museum], Stanford University, no. 10, 4 pp. June 15. 

*563. Generic classification of the high-altitude pelobatid toads of Asia {Scutiger, Aeluro- 
phryne, and Oreolalax). (G. S. Myers and Alan Edward Leviton.) CO, 1962, no. 2, 
pp. 287-291. July 20. 

1963 

564. KilHfish identification. American KilUfish Association, Killie Notes, Vol. 2, no. 2, 

pp. 7-10. March. 

565. Fresh-water fishes. Pacific Discovery, San Francisco, vol. 16, no. 4, pp. 36-39. July. 

[Freshwater fishes of the world, general facts, diversity, sizes, distribution, con- 
servation.] 

566. Comments on the proposed rejection of the type designations of Jordan and Evermann 

1896-1900 and 1896. Bulletin of Zoological Nomenclature, vol. 20, part 4, p. 259. 
July 12. [See also 561.] 

567. The fresh-water fauna of North America. Proceedings, XVI International Congress 

of Zoology, vol. 4, pp. 15-20. August. [Only abstracts were published in this volume. 



Vol. XXXVIII] MYERS: BIBLIOGRAPHY 51 

The paper as delivered at the Congress was much longer and argued in favor of 
continental drift.] 
568. Foreward. In: Jordan, D. S., The genera of fishes and a classification of fishes (re- 
print; Stanford University Press; xvi + 800 pp.), pp. vii-xvi. December 30. [Sets 
these two important works in perspective, gives considerable historical information 
on zoological nomenclature and fish classification, and warns against misuse of the 
Jordan papers.] 

1964 

*569. An electrophoretic survey of rattlesnake venoms. (Alan E. Leviton, G. S. M^-ers, and 
B. W. Grunbaum.) In: Leone, C. A. (editor), Taxonomic biochemistry and serology 
(New York; Ronald Press: x + 728 pp.), pp. 667-671. [On the basis of the venoms 
of 10 species, relations similar to those shown by morphology were found.] 

570. A brief sketch of the history of ichthyology in America to the year 1850. CO, 1964, 

no. 1, pp. 33-40. March 26. 

571. Foreward. In: Albert W. Herre (1868-1962): a brief autobiography (Division of 

Systematic Biology, Stanford University, Circular no. 10, 20 pp.), pp. 1-2. March. 

572. Bumblebee catfishes {Plotosus) . TFH, vol. 13, no. 3, pp. 5-7, 75. November. [They 

are black-and-j^ellow, they buzz, they swarm, and they sting.] 

1965 

573. Gambusia, the fish destroyer. TFH, vol. 13, no. 5, pp. 31-32, 53-54. January. [First 

indictment of Gambusia affinis, the mosquitofish, as a serious danger to other fishes, 
small and large, wherever introduced. See also 575.] 

574. The body-wag, an innate behavioral characteristic of bony fishes. TFH, vol. 13, no. 9, 

pp. 21, 24-25. May. 

575. Gambusia, the fish destroyer. Austrahan Zoologist, vol. 13, no. 2, p. 102. August. 

[Partial reprint of 573, with editorial note about Australian introductions.] 

576. Races of the Chinese paradise fish {Macropodus) . TFH, vol. 14, no. 1, pp. 48-49. 

[Continuation of taxonomic revision begun in 137. Southernmost race in Indochina 
is M . opercularis concolor Ahl. Adds Polyacanthus yangye Dabry 1872 to synonymy 
of M. chinensis.] 

1966 

577. Foreword [to first issue]. Ichthyologica, the Aquarium Journal [continuation of AJ], 

vol. 37, no. 1, pp. 3-5. January. [G.S.M. resigned as editor after this issue appeared.] 

578. How to become an ichthyologist. Part 1. TFH, vol. 14, no. 8, pp. 47, 50-51. April. 

[See 580, 581, 583. Advice for young prospective ichthyologists.] 

579. Phyletic studies of teleostean fishes, with a provisional classification of living forms. 

(P. Humphry Greenwood, Donn Eric Rosen, Stanley Howard Weitzman, and G. S. 
Myers.) Bulletin of the American Museum of Natural History, vol. 131, art. 4, 
pp. 339-446, pis. 21-23. April 18. 

580. How to become an ichthyologist. Part 2. TFH, vol. 14, no. 9, pp. 47, 50-51. May. 

[See also 578, 581, 583.] 

581. How to become an ichthyologist. Part 3. TFH, vol. 14, no. 10, pp. 28-30. June. 

[See also 578, 580, 583.] 



52 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

582. Two remarkable new trichomycterid catfishes from the Amazon basin in Brazil and 

Colombia. (G. S. Myers and Stanley Howard Weitzman.) Journal of Zoology [con- 
tinuation of the Proceedings of the Zoological Society of London], vol. 149, pp. 
277-287. July. 

583. How to become an ichthyologist. Part 4. TFH, vol. 14, no. 2, pp. 29-31. August. 

[See also 578, 580, 581.] 

584. Megalomycteridae, a previously unrecognized family of deep-sea cetomimiform fishes 

based on two new genera from the North Atlantic. (G. S. Myers and Warren Curtis 
Freihofer.) SIB, vol. 8, no. 3, pp. 193-206. October 7. 

585. Derivation of the freshwater fish fauna of Central America. CO, 1966, no. 4, pp. 

766-773. December 23. [Paper read before the American Society of Ichthyologists 
and Hcrpetologists in June 1964.] 

1967 

586. [Review of] Breder, C. M., and Rosen, D. E., Modes of reproduction in fishes. Natural 

History, N.Y., vol. 76, no. 2, pp. 66-67. February. [Suggests possibly primitive 
nature of nesting and parental care in bony fishes and notes its widespread oc- 
currence.] 

587. Note on the name of a Guatemalan cactus, MammHlaria voburnensis Scheer. Cactus 

and Succulent Journal of America, vol. 39, no. 4, p. 153. July. 

588. Zoogeographical evidence of the age of the South Atlantic Ocean. Studies in Tropical 

Oceanography, no. 5 (Miami, xx + 847 pp.), pp. 614-621. October 1. [Paper read 
at International Conference on Tropical Oceanography, Miami, two years previously.] 

589. Named main divisions of teleostean fishes. (P. Humphry Greenwood, G. S. Myers, 

Donn Eric Rosen, and Stanley Howard Weitzman.) PBSW, vol. 80, pp. 227-228. 
December 1. 

590. Note on the dentition of Creagnidite niaxillayis, a characid fish from the upper 

Orinoco-upper Rio Negro system. (G. S. Myers and Tyson Royal Roberts.) SIB, 
vol. 8, no. 4, pp. 248-249. December 5. 

1969 

591. The endemic fish fauna of Lake Lanao, and the evolution of higher taxonomic cate- 

gories. /«: Ehrlich, P. R., Holm, R. W., and Raven, P. H. (editors), Papers on 
evolution (Boston, Little Brown and Co., xii + 564 pp.), pp. 247-261. [Reprint of 
537, with corrections.] 

592. [Review of] Harden Jones, F. R., Fish migration. CO, 1969, no. 2, pp. 409-411. 

June 3. 

593. Peace! It's wonderful! [Allegory.] Palo Alto Times, Thursday, November 27, 1969, 

p. 26. [Based on a true story.] 

594. The endemic fish fauna of Lake Lanao, and the evolution of higher taxonomic cate- 

gories. In: Laetsch, Watson M., the biological perspective (Boston; Little, Brown 
and Co., xii -f- 574 pp.), pp. 351-365. [There were two reprints of this paper during 
1969, this being the second. Sec also 591. Both are reprints of 537.] 



PROCEEDINGS 

OF THE 

CALIFORNIA ACADEMY OF SCIENCES 

FOURTH SERIES 



Festschrift for George Sprague Myers 



Vol. XXXVIII, No. 3, pp. 53-62; 3 figs.; 1 table. December 31, 1970 



A NEW SPECIES OF 

THE DORADID CATFISH GENUS 

LEPTODORAS, 

WITH COMMENTS ON RELATED FORMS 

By 

James E. Bohlke 

Chaplin Chair oj Ichthyology 

Academy of Natural Sciences of Philadelphia 

Abstract: Leptodoras myersi is described from a trawl haul made in Rio Amazonas 
near Iquitos, Peru. Leptodoras juriiensis, previously known only from the holotype 
taken in Rio Jurua, Brasil, is recorded from the same haul. 

It was originally intended that this be a revision of the genus Leptodoras but 
at the last moment it became apparent that the species L. linnelli is a composite 
that will require further study. Also, in Eigenmann's (1925) review of the 
Doradidae, the related genera Opsodoras, Hassar, and Leptodoras are perhaps 
the least well defined and thus require more attention. At present, I describe as 
new a well-marked species and comment on its relatives. 

The newly recorded specimens of Leptodoras myersi and L. juruensis were 
collected on the 1955 Catherwood Foundation Peruvian-Amazon Expedition by 
Charles C. G. Chaplin and Ruth Patrick of the Academy's staff. They were 
taken with an otter trawl from the Amazonas (Maraiion) between Isla Iquitos 
and Isla Lapuna. Only one downstream haul was made, because of the swiftness 

[S3] 



54 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

of the current and the many snags in the bottom, but this caught a fascinating 
group of mostly new and rare catfishes plus one specimen of Rhytiodus microlepis 
Kner. This suggests that more bottom sampling should be attempted in the 
large South American rivers. 

For the loan of important specimens, I thank P. H. Greenwood of the British 
Museum (Natural History) and W. I. Follett of the California Academy of 
Sciences. 

METHODS 

The standard length measurement was made with some difficulty but, by 
flexing the caudal fin and using reflected light, I believe fairly good accuracy 
was achieved. The head length includes the fleshy opercular flap. The eye and 
snout measurements involve the eyeball proper. The predorsal measurement, 
length of dorsal spine, and depth at dorsal-fin origin all have as one terminus the 
anterior groove of the small bony element at the base of the spine. The anterior 
end of the adipose dorsal-fin base is somewhat difficult to define, but measure- 
ments involving it are more easily duplicated in this species than in L. acipen- 
serinus for example. The greatest width of the peduncle is at the posterior end 
of the anal-fin base and includes the lateral spines that jut out from the body. 
The length of the pectoral spine is measured basally from the notch, where a 
needle-point fits in snugly, and not from the extreme base of the spine. Ventral- 
fin length is the greatest length of the fin, not of an individual ray. The greatest 
head width is of the bony portion and not of the rather indifferent fleshy portion 
posteriorly. The greatest scute depth is a vertical measurement of the area 
covered by scutes, rather than the diagonal measurement of an individual scute. 

The anal-ray counts are separated into anterior unbranched, small roman 
numerals, and posterior branched, arable numerals; the last anal ray some- 
times is simple, sometimes consists of two rays united at their bases, but in either 
case is counted as one. The lateral scutes are all of those in the main lateral 
series, including the ones on the caudal-fin base but not the small ones in the 
humeral region, the tympanum. 

Leptodoras myersi Bohlke, new species. 

Diagnosis. This is an elongate, long-snouted species like L. acipenserinus, L. 
Itnnelli, and L. juruensis. Its dark color markings, particularly the broad nuchal 
band, are distinctive. Lateral scutes few, 36 or 37, each scute bearing few points. 
Total anal-fin rays few, 13 or 14 (except see discussion of L. linnelli below). 
Lacking extremely elongate first dorsal spine of L. juruensis. While anterior 
base of adipose dorsal fin is not sharply defined, it does not extend far forward 
as a fleshy ridge. Dorsal and pectoral spines with small hooks or spines along 
their anterior and posterior margins, these weakest on the dorsal spine and 
strongest on the posterior margins of the pectoral spines. Nuchal foramen 



Vol. XXXVIII] BOHLKE: NEW SPECIES OF LEPTODORAS 



55 




Figure 1. Leptodoras niyersi: Holotype, 74.6 mm. standard length, ANSP 112318. 



present. Head covered by small, elongate, pale fleshy ridges, arranged in a 
pattern (see fig. 1). 

Description. The body shape and dark color markings are shown on the 
photographs (fig. 1). Selected measurements and counts made on a series of 
10 specimens appear in table 1. 

Dorsal rays I, 6. Total anal rays 13 or 14, nearly always 14. Pectoral rays 
I, 9 or I, 10, usually I, 10. Ventral rays i, 6/i, 6. Principal caudal rays i, IS, i, the 
ventral unbranched ray counted not extending back to the tip of the lobe as does 
the dorsal one. Number of lateral scutes 36 or 37 in equal numbers. Anterior 
dorsal serrae 7 to 12, posterior dorsal serrae 6 to 11, anterior pectoral serrae 19 
to 23, and posterior pectoral serrae 12 to 15; the numbers of serrae apparently 
are not related to the length of the fish, at least within the limited size range 
examined. 

No teeth present. Nostrils both with raised margins, that of the rear nostril 
lowest posteroventrally. Anterior nostril nearer eye than tip of snout. Distance 
between the two nostrils on one side equal to that between posterior nostril and 



56 



CALIFORNIA ACADEMY OF SCIENCES 



[Proc. 4th Ser. 



a 
















"^ 


oo 


!^ 


CO 


ro 


OO 


0^ 


•* 


« 


^ 


CO 


CO 


o 


1^1 

o 


O 





J>! -"^ o> 
t^ ro --I O 



O l» 00 00 1--; O 
-* ro Tt" O ro t^ ro 



O 00 mD »— j CM 00 lO lO 



vOrooO'^f^l^OC^fNCO^-t^'— ' 



Tf fv] 1— I Q\ VO 



oO"— lOOOrof^^iOO"— icsiT-Hi-i'^ 



r^vOvDOCT^'— lOCOO^OTf 



vO 


Ov 


C'O 


o 






o> 


o 



r^l ro "^ rN ^ O O 

■^ O --^ CO ro C^ '— I — ^ «„ 

O^'^OO'— 'c^l"— '■— I 



(Vj CO 

CO CO i<-j 



O i>i rf 



•<i; vq ro c~) 
C> ^ ON Tf 
MO rO "-I O 



rs) ^ O (^1 J^ ""O C 
O O ■* <~^i O O ■- 



00 O rO --H 00 
00 lO O -^ rs 
VO ro C<| O O 



^ "* ■"! 
On ^' "^ 
-1 O O 



t^ <?}- rO On r^ ^ i~^ O '-H ro *^_ 

I— iTfoOr<-)ONroCOCOMDONi^ 



rv)O^ONONCOO;CO'^<X3t~Nl'-;'~^ 



O •* i^l O 



Ti- r^ ^ O 



lO<^'^r<lONf^lOvOOOOCO'— 'rfoq-^C^rotvj 
CO'^ONrJ-rvi'^Oc-lt^fOON^ONCOloO^O^ 
vOrO'-iOOO-^tNlOO'- 



ONNOO<^ONr<l^*^ 



"-) rO rO O r-) -^ 



I^ Tf CO ^ 



Tf O <~0 t^ 



NOrOi-lOOO-^rMO 



r<icONOO^cO\0'-HONO'+'-' 



r-H 00 !^1 O 



C j3 ^ 



^ s ^ 

-5 j: oj „ 



- « -g ^ •" 

03 o ■" IJ C 






c 


















G 


43 


























4= 
















bu 


a 

X) 

G 


o 
<u 
O 


"5 


43 


42 


til 




43 


u 

C 


ID 





C 


U 






61 1 




45 




f) 




M-i 








CD 


c 


<D 

G 

'S. 
tn 


G 
(1) 


5: 


n 




to 


42 


G 


■a 

a 


a 

3 


G 

"3. 


_G 


43 


13 


.s 


-O 





'i~i 


,^_j 






. 


'i* 


, I 


;-4 


1 


03 


t . 


































(fl 


O) 


1-i 




43 


<L1 










O 


Lh 






G 




:-< 


a 




^ 


G 




> 


rt 


D. 


03 




^ 








0) 












u 


O 


hJ 


Q 


CL, 


> 





Ph 


Oh 


U 






Vol. XXXVIII] BOHLKE: NEW SPECIES OF LEPTODORAS 



57 






Figure 2. Leptodoras myersi: Paratype, 63.3 mm. standard length, AXSP 112320. A, 
outline of sixth lateral scute on left side of fish (central ridge not indicated), 3.1 mm. in 
longest dimension; B, dorsal spine with break-away tip, main spine 11.5 mm.; C, right 
pectoral spine, 14.2 mm. 



eye. Eyes distinctly horizontally elongate. Horizontal width of eye equal to 
least width of bony interorbit, or nearly so. 

The middorsal fontanel is long and narrow, extending from just behind a line 
drawn between the rear margins of the posterior nostrils to one drawn between 
the rear margins of the eyes. A middorsal groove extends in front of the fontanel 
on the snout and posteriorly from the fontanel to the dorsal-fin origin. Gill 
membranes connected to isthmus ventrally, the ventral ends of the two gill open- 
ings separated by a space of slightly more than the least interorbital width. Gill 
rakers low fleshy bumps. Two m.axillary and 4 mental barbels present, all inter- 
connected to form a hood of considerable size when spread. Each maxillary 
barbel is divided, its outer portion with barblets along its outer margin; these 
barblets are in two series, the upper ones short and numerous, the lower ones 
long and fewer in number. The outer portion of the maxillary barbel usually 
fails to reach the ventral end of the gill opening but may just attain that level. 
The inner portion of the maxillary barbel forms the lateral margin of an ex- 
tensive fleshy lobe that connects the maxillary with the outer mental barbel; 
this lobe has short barblets around its margin. The mental barbels are studded 
with short barblets all around. The maxillary is ossified to the extent of about 
one-third the length of the outer division of the barbel. Branchiostegal rays 7. 
Upper end of gill opening just behind lower end of upper third of dark opercular 
spot shown on figure 1. 

On the dorsal spine, the anterior serrae are longer than the posterior; the 
anterior ones are crowded and overlapping, especially basally, are directed toward 
the spine tip, and are distributed along less than the basal half of the spine. The 
posterior serrae are short, widely spaced, their tips directed perpendicular to the 
main axis of the spine or slightly inclined either basally or distally; when their 



58 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

tips are angled, the outer serrae are tilted basally and the lower serrae are tilted 
distally. On one individual the uppermost serration is directed distally. The 
posterior serrae extend along the distal one-half to three-quarters of the spine. 
There is a distinct ossified break-away tip on the spine when intact, this tip 
fitting into a median groove in the spine proper; the dorsal spine outlined in 
figure 2 shows this tip nearly completely disengaged. 

On the pectoral spines the anterior serrae are shorter than the posterior, are 
directed toward the spine tips, and the distalmost one counted forms the tip of 
each spine (not the case in several related species in which the spine tips are 
rounded or blunt and without projections). The posterior serrations are strong 
hooks, their tips directed toward the bases of the spines. There are pointed 
fleshy extensions beyond the ossified spine tips, indicating that there probably 
are no ossified break-away segments as on the dorsal spine of this species and the 
pectoral spines of L. jiiruensis (the latter will be discussed and figured in a sub- 
sequent paper when the status of L. linneUi also will be treated). Pectoral spines, 
when depressed, extending back well beyond the ventral-fin bases. A single 
pectoral pore present on each side. 

Tips of ventral fins rounded, not extending back to the anal-fin origin. Anal 
and genital papillae placed between the ventral fins, at or slightly before mid-fin. 
Adipose dorsal fin well developed, short-based, its origin above points varying 
between the base of the sixth to the interspace between the seventh and eighth 
anal rays. Caudal fin distinctly forked, the lower lobe the longer. On the single 
stained individual there are 18 procurrent caudal rays above and 17 below the 
principal rays. 

In the humeral region there is a sharp spine on the posterior margin of the 
supraclavicle followed, across the center of the tympanum, by 3 narrow, elongate, 
mostly embedded ossifications. The first of the three is longest and has a rise 
in the middle that appears as a low hump externally; the second is completely 
embedded or with only a minute portion exposed; the third is shortest and bears 
a sharp projecting spine at its posterior end. The pointed coracoid processes ex- 
tend posteriorly beyond the base of the last pectoral ray, but not as far back as 
do the humeral processes; the last-mentioned are shallowly convex dorsally and 
terminate in a narrowly rounded tip. There is a horizontally elongate nuchal 
foramen on each side. 

The lateral series of scutes begins behind the tip of the humeral process and 
continues out onto the caudal fin basally. The scute outlined in figure 2 is much 
like those just before and after it (the outline does not indicate the median 
longitudinal ridge and the sharp outward angulation of the median spine itself) ; 
proceeding posteriorly the scutes become increasingly wider, less deep, less 
angular, and more overlapping. The terminal ossification in the series is a nar- 
row, elongate, tubular element that lacks a spine and usually is overlooked on 



Vol. XXXVIII] BOHLKE: SEW SPECIES OF LEPTODORAS 59 

unstained specimens. This was not included in the counts of scutes recorded in 
this paper; apparently it also was not counted by Eigenmann (1925: 358), for 
I obtained the same count as he did on the same specimen of L. acipenserinus, 
omitting this element. 

The pattern of coloration in alcohol is shown in figure 1 ; however, the broad 
nuchal band is more definitely continuous across the dorsum than is suggested 
on the dorsal view, and the dark opercular margin frequently is more intense 
and continues farther ventrally. A faint dusky stripe is present on the upper 
half of the lower caudal-fin lobe, and sometimes there is an even fainter one on 
the lower half of the upper lobe. The dorsum is dusky between the dorsal fin and 
caudal-fin base. The top of the head is dusky before the nostrils and in a roughly 
circular middorsal patch immediately behind the eyes. The basal half or more 
of the pectoral fin exclusive of the spine (except sometimes the membrane en- 
casing the posterior serrae) usually is distinctly dark, sometimes only dusky and 
the extent of the dusky area variously more reduced than shown in the figure. 

Relationships. The elongate, long-snouted species L. acipenserinus, L. 
Unnelli (both types if they represent more than one species) and L. juruensis 
are most closely related to L. mycrsi. Leptodoras juruensis is the most distinctive 
and most spectacular looking member of this group, with its extremely elongate 
anterior dorsal-fin element and distinctive black color markings (see fig. 3); it 
also has more lateral scutes than the others: 44 to 46 in L. juruensis, 42 in L. 
acipenserinus, 38 or 39 in the L. Unnelli complex, and 36 or 37 in L. myersi. 
Leptodoras juruensis is distinctive in certain proportions, but these will be treated 
later. Leptodoras mycrsi has a lower anal-fin ray count than other species of 
Leptodoras excepting, evidently, the Guianan (typical) population of the L. 
Unnelli complex. Eigenmann (1912: 192), in his original description of L. 
Unnelli, gave an anal-ray count of 12 to 14; the specimen (ANSP 39734) from 
Rio Rupununi recorded by Fowler (1914: 264) has 12 rays. Leptodoras myersi 
has 13 or 14 rays, L. acipenserinus has 17 rays (16 recorded for the holotype by 
Gunther 1868: 230), L. juruensis has 16 or 17 rays, and the Peruvian specimens 
nearest L. Unnelli have 15 to 17 rays. 

In the size range represented by the Peruvian material, the shapes of the 
lateral scutes are most similar in L. myersi and L. juruensis, those of L. acipen- 
serinus and nominal L. Unnelli having more teeth above and below the median 
spine. 

While differing in numerous relative proportions (to be discussed further 
when the L. Unnelli question is resolved), L. myersi has a distinctly smaller eye 
than L. Unnelli (four or five percent of standard length versus seven or eight 
percent). Leptodoras Unnelli has no color pattern except for a faint dusky stripe 
that extends out each caudal lobe, while L. mycrsi has the distinctive pattern 
described above. Leptodoras acipenserinus is described as devoid of a color 



60 



C A LI FORM A ACADEMY OF SCIENCES 



[Proc. 4th Ser. 




Figure 3. Leptodoras juruensis: 125.8 mm. standard length, ANSP 112321. 



pattern (Giinther 1868: 230) and lUM 15878 (Eigenmann 1925: 358) shows 
no trace of one. This color difference, coupled with the differences listed above 
between the elongate L. acipenserinus and L. myersi, plus numerous proportional 
differences that will be outlined in a subsequent paper, indicate how different 
are the two species. Leptodoras myersi has elongate, pale, raised ridges on the 
head, L. acipenserinus has pale low papillae, while L. Unnelli has nothing of the 
sort. 

Name. For my professor and good friend, George S. Myers. 

Material EXAMINED. Holotype: ANSP 112318 (74.6 mm. standard length, 
photographed), Peru: vicinity of Iquitos, Rio Amazonas (Marafion) between 
Isla Iquitos and Tsla Lapuna, near Isla Lapuna shore; to 12 ft. (3.66 meters); 
trawl; 9 October 1955; C.C.G. Chaplin, R. Patrick. 



Vol. XXXVIII] BOHLKE: NEW SPECIES OF LEPTODORAS 61 

Paratypcs: ANSP 112319 (9; 54.9-77.6), ANSP 112320 (1; 63.3, cleared 
and stained) and USNM 203816 (2; 68.5-74.8), taken with the holotype. 

Leptodoras juruensis Boulenger. 

Leptodoras juruensis Boulenger 1S98, p. 478 (Type locality: Rio Jurua, Brasil). Eigen- 
MANN 1925, pp. 357, 358 (diagnosis based on type). 

Previously known only from the holotype from Rio Jurua, this species now 
is recorded from the same trawl haul that collected the type material of L. 
myersi. A fine Peruvian example is illustrated in figure 3. Peruvian specimens 
have been compared with the much larger holotype of the species and the results 
of this comparison will be forthcoming. 

Material examined. British Museum (Natural History) 1898-10-11-25 
(223 mm. standard length, holotype), Brasil: Rio Jurua; Goeldi. ANSP 112321 
(1; 125.8, photographed), ANSP 112322 (6; 71.4-96.7) and USNM 203817 
(1 ; 92.6), taken with the holotype of L. myersi. 

NOTE 

While this manuscript was in proof, a paper on Venezuelan doradids was received from 
Fernandez Yepez (1968, Boletin del Instituto Oceanographico, Universidad de Oriente, 
Cumana, Venezuela, vol. 7, no. 1, pp. 7-72). In it, he includes the species "leporhinus," 
'Hinnelli," and ^^notospilus" in the genus Opsodoras, whereas "linnelli" previously was in 
Leptodoras and "notospilus" was in Hassar. Subsequent correspondence with that author 
revealed that his rationale for making these and other nomenclatural changes is in a manu- 
script still in press. The new species, "myersi," is closest to "linnelli" and "acipenserinus," 
which were placed by Eigenmann (1925: 357) in Leptodoras, so the name combination 
Leptodoras myersi is here published, with the realization that the species may later be 
transferred to a different genus. 

REFERENCES 
Boulenger, G. A. 

1898. Descriptions of two new siluroid fishes from Brazil. Annals and Magazine of 
Natural History, ser. 7, vol. 2, pp. 477-478. 

ElGENMANN, C. H. 

1912. The freshwater fishes of British Guiana, including a study of the ecological group- 
ing of species and the relation of the fauna of the plateau to that of the low- 
lands. Memoirs of the Carnegie Museum, vol. 5, xxii + 578 pp., 103 pis. 

1925. A review of the Doradidae, a family of South American Nematognathi, or cat- 
fishes. Transactions of the American Philosophical Society, new series, vol. 22, 
no. 5, pp. 28Q-365, 27 pis. 
Fowler, H. W. 

1914. Fishes from the Rupununi River, British Guiana. Proceedings of the Academy 
of Natural Sciences of Philadelphia, vol. 66, pp. 229-284. 

GUNTHER, A. 

1868. Descriptions of freshwater fishes from Surinam and Brazil. Proceedings of the 
Zoological Society of London, 1868, pp. 229-247, pis. 20-22. 



PROCEEDINGS 

OF THE 

CALIFORNIA ACADEMY OF SCIENCES 

FOURTH SERIES 



Festschrift for George Sprague Myers 



Vol. XXXVIII, No. 4, pp. 63-98; 8 figs. December 31, 1970 



SYSTEMATICS OF THE GENUS HEMITRIAKIS 

(SELACHII: CARCHARHINIDAE), AND 

RELATED GENERA 

By 

L. J. V. Compagno 
Division of Systematic Biology, Stanford University 

INTRODUCTION 

Herre (1923) described Hemitriakis leticoperiptera, a new genus and species 
of shark from the Philippine Islands. Hemitriakis was thought to differ from 
Triakis Miiller and Henle in its dentition, snout, nasal valves, body, and caudal 
fin. However, Fowler (1941), Bigelow and Schroeder (1948), Garrick (1954), 
and Kato (1968) considered Hemitriakis a junior synonym of Triakis. 

Present data shows that Hemitriakis is a well defined genus with two species: 
H. leucoperiptera Herre, 1923; and H. japanica (Miiller and Henle, 1841). This 
account is a review of the systematics of Hemitriakis and related genera in the 
family Carcharhinidae. 

ACKNOWLEDGMENTS 
Reeve M. Bailey (University of Michigan Museum of Zoology), Myvanwy 
M. Dick (Museum of Comparative Zoology, Harvard University), W. I. Fol- 
lett, William D. Eschmeyer, Lillian J. Dempster (Department of Ichthyology, 
California Academy of Sciences), and Stanley H. Weitzman (Division of Fishes, 
U. S. National Museum) loaned or gave specimens to me and provided working 
facilities at their institutions. C. G. Alexander (Department of Biology, San 

[63] 



64 



CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 







Figure 1. A, dorsal view, and D, ventral view, of head of Hemitriakis japanica 
(SU-12677). B, dorsal view of head of Hypogaleus hyugaensis, adopted from Miyosi 
(1939). C, dorsal view, and F, ventral view, of head of Galeorhiniis zyopterus (LJVC-0238; 
847 mm. female.). E, ventral view of head of Hypogaleus zanzibariensis, adopted from 
Smith (19S7b). Abbreviations: HHR, horizontal head rim; SR, subocular ridge. 



Vol. XXXVIII] COMPAGXO: SYSTEMATICS OF GENUS HEMITRIAKIS 65 

Francisco State College), Robert P. Dempster (Steinhart Aquarium, California 
Academy of Sciences), and Louis Garibaldi (American Broadcasting Company 
Marine World, Redwood City, California) supplied many fresh and frozen car- 
charhinids for anatomical preparations. The late J. L. B. Smith (Department of 
Ichthyology, Rhodes University, Grahamstown, South Africa) sent specimens of 
Eridacnis sinuans and Scylliogaleus quecketti; Leslie W. Knapp (Smithsonian 
Oceanographic Sorting Center, Washington, D. C), C. Richard Robins, and 
Phillip C. Heemstra (Institute of Marine Sciences, L^niversity of Miami) loaned 
other carcharhinids. In addition to providing numerous specimens and research 
facilities, Shelton P. Applegate (Division of Vertebrate Palaentology, Los An- 
geles County IVIuseum of Natural History), Susumu Kato (Bureau of Commer- 
cial Fisheries Fishery-Oceanography Center, La Jolla, California), and Stewart 
Springer (Bureau of Commercial Fisheries Systematics Laboratory, U. S. Na- 
tional INIuseum) have discussed various aspects of carcharhinid taxonomy cov- 
ered in this paper with me. J. A. F. Garrick (Department of Zoology, Victoria 
University of Wellington, New Zealand) sent comments on several systematic 
problems concerning carcharhinid genera and species. George S. Myers critically 
reviewed the first draft of the manuscript, and Warren C. Freihofer (Division of 
Systematic Biology, Stanford University) offered useful suggestions. I am most 
grateful for the help offered by all of these people, without which this account 
could not have been written. 

STUDY MATERIAL 

Specimens mentioned in the text and figures are from the collections of the 
George Vanderbilt Foundation at the California Academy of Sciences (GVF); 
Division of Systematic Biology, Stanford University (SU) ; University of Michi- 
gan INIuseum of Zoology (UMMZ) ; U. S. National Museum (USNM); and of 
the writer (LJVC). 

Hemitriakis specimens examined are listed below, with number of specimens 
and total lengths in parentheses. 

Hemitriakis japanica: SU-1 267 7, Nagasaki, Japan (1; 682 mm.); UMMZ- 
179060, Auraji (Osaki Market, Osaki), Japan (1; 650 mm.); UMMZ^179061, 
Ainoshima (Fukuoka Market, Fukuoka), Japan (1; 560 mm.); UMJ\IZ-1 79062, 
Ezumi (Ezumi Market), Japan (1; 505 mm.); USNM-191193, Taipeihsien, 
Taiwan (3; 651-685 mm.). 

Hemitriakis leucoperiptera: SL)-27118, Dumaguete, Oriental Negros, Philip- 
pine Islands (2; 169-170 mm.). 

Hemitriakis species: SU-40097, Dumaguete, Oriental Negros, Philippine 
Islands (4; 161-180 mm.). 

Comparative material including most carcharhinid genera and species was 
examined. As the number of specimens in this sample is enormous, they are 
not listed here but will be given in a forthcoming revision of carcharhinid genera. 



66 



CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 






© 



UJ 


CL 


-J 


o 


CO 


(/) 




Vol. XXXVIII] COMPAGNO: SYSTEMATICS OF GENUS HEMITRIAKIS 67 

Instead, the genera and species examined are listed. Carcharhinus species no- 
menclature is modified from Garrick (1967); that for Scoliodon, Rhizoprion- 
odon, and Loxodon is from V. Springer (1964). 

Aprionodon isodon, Carcharhinus acronotus, C. albimarginatns, C. altimus, C. 
amblyrhynchus, C. borneensis, C. cauta, C. falciformis, C. galapagensis , C. leucas, 
C. limbatus, C. longimanus, C. macuUpinnis, C. melanopterus, C. menisorrah, C. 
milberti, C. obscurus, C. pleurotaenia, C. porosus, C. remotus, C. sorrah, C. 
springeri, C. tjutjot, C. velox, Eridacnis barbouri, E. radcliffei, E. sinuans, Galeo- 
cerdo cuvier, Galeorhinus australis, G. chilensis, G. galeus, "G." omanensis, G. 
zyopterus, Hemigaleus baljowi, H. macrostoma, H. microstoma, H. pectoralis, 
H. tengi, Hemipristis elongatus, Hypoprion hemiodon, H. macloti, H. signata, 
Isogomphodon oxyrhynchus, Lamiopsis temmincki, Leptocharias smithii, Loxodon 
macrorhinus, Mustelus antarcticus , M. asterias, M. californicus, M. cams, M. 
dorsalis, M. jasciatus, M. griseus, M. henlei, M. higmani, M. kanekonis, M. 
lenticulatus, M. lunulatus, M. manazo, M. mento, M. mustelus, M. norrisi, M. 
schmitti, Negaprion acutidens, N . brevirostris, iV. jorsteri, A^ jronto, Prionace 
glattca, Proscyllium habereri, Rhizoprionodon acutus, R. lalandei, R. longurio, R. 
oligolinx, R. porosus, R. terraenovae, Scoliodon laticaudus, Scylliogaleus 
quecketti, Triaenodon obesus, Triakis acutipinna, "T." jehlmanni, T. maculata, 
T. scyllia, T. semifasciata. 

TERMINOLOGY 

For descriptive purposes the morphological terminology of the head, eyes, 
dentition, vertebral column, and fins of carcharhinid sharks is discussed and 
elaborated here. 

Head Morphology. The horizontal head rim (fig. 1) is the head margin in 
dorsal or ventral view. The subocular ridge is a ventrolateral expansion of the 
horizontal head rim beneath the eye. In Hemitriakis, Triakis, Mustelus, Fur- 
galeus, and other carcharhinid genera with well developed subocular ridges, the 
eyes appear medial to the horizontal head rim in dorsal view. A subocular ridge 
obscures the eyes in ventral view. 

Nictitating Lower Eyelid (fig. 2). Form and terminology of the carchar- 
hinoid ocular structures variously termed nictitating membranes, nictitating 



Figure 2. Lateral views of carcharhinid eyes, showing nictitating lower eyelid types. A. 
Proscyllium habereri (UMMZ-1 79064 ; S3S mm. female), with rudimentary NLE. B. Mustelus 
cants (USNM-197676; 337 mm. female), with external NLE. C. Galeorhinus australis 
(USNM-17699S; 385 mm. male), with transitional NLE. D. Leptocharias smithii (USNM- 
202677; 570 mm. male), with internal NLE. Abbreviations: NLE, nictitating lower eyelid; 
SLE, secondary lower eyelid; SOP, subocular pouch; SP, spiracle; UE, upper eyelid. 
Dashed hne is bottom of subocular pouch; dotted line in Leptocharias is edge of NLE 
inside palpebral aperture. 



CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 



® 




PML 



PLC 



PLL 




PLC 



TN BG 



Figure 3. Anteroposterior teeth of Hemitriakis japanica (SU-12677). A. Inner face 
of right lower tooth. B. Outer face of left lower tooth. Abbreviations: BG, basal groove; 
BL, basal ledge; CR, crown; FT, crown foot; PC, primary cusp; PLAS and PMAS, 



Vol. XXXVIII] COMPAGXO: SYSTEMATICS OF GENUS HEMITRIAKIS 69 

folds, subocular folds, movable lower eyelids, and nictitans were reviewed by 
Gilbert (1963) and by Gilbert and Oren (1964). They used the term "nictitans" 
to cover all variations of the mobile eyelid of scyliorhinids and carcharhinids, but 
this term is not adopted here as the selachian structure is morphologically and 
developmentally unlike the true nictitans or nictitating membrane of tetrapods 
and in many cases is merely a little-modified movable lower eyelid. Instead, the 
term nictitating lower eyelid (NLE) is introduced to avoid some of the connota- 
tions of nictitans and to recognize the probable derivation of the structure from 
the original lower eyelid of precarcharhinoid sharks. 

The exterior fold formed by the groove below the NLE is termed the second- 
ary lower eyelid (SLE). The groove itself is the subocular pouch. 

Four nictitating lower eyelid types can be distinguished among carcharhinids 
if subdivisions are made in the morphological gradient seen in this structure. The 
rudimentary type is the least specialized. In it the NLE forms the ventral edge 
of the palpebral aperture and connects anteriorly and posteriorly with the upper 
eyelid. The SLE is a weak ridge below the NLE and does not connect with either 
the upper eyelid or the NLE. The upper edge of the SLE is not defined and the 
subocular pouch is a very shallow, external groove. The external type differs 
from the rudimentary in that the SLE is a strong flap with a well defined edge. 
The subocular pouch, although relatively shallow, is strongly differentiated. The 
internal type is the most advanced, with the NLE ends entirely internal to the 
palpebral aperture and not connected to the upper eyelid. The SLE replaces the 
NLE in contacting the anterior and posterior ends of the upper eyelid and forms 
the ventral edge of the palpebral aperture. The subocular pouch is entirely within 
the palpebral aperture and varies from moderately shallow (Leptocharias) to 
very deep {Carcharhinus). The transitional type covers intermediates between 
internal and external types. These often have the SLE attached by one of its 
ends (posterior or anterior) to the upper eyelid, while the NLE has its opposite 
end also attached to the upper eyelid. 

Dentition (figs. 3-4). Tooth topography of selachians was discussed briefly 
by Applegate (1967). He divides the tooth into two external regions, the crown 
and the root. The crown is the enamel-covered region of the tooth distal to its 
attachment with the jaw. The proximal root lacks the enamel covering and has 
its component osteodentine exposed to the surface. The region of the crown 
proximal to the root is termed the foot. As used by Bigelow and Schroeder 
(1948), the term base includes both the root and the crown foot. 

The crown and root are both compressed in a plane with its horizontal sides 



postlateral and premedial parts of attachment surface; PLC, postlateral cusplets; PLL, 
postlateral lobe of root; PME, premedial edge of crown; PML, premedial lobe of root; 
RT, root ; TG, transverse groove ; TX, transverse notch ; TR, transverse ridges. 



70 



CALIFORNIA ACADEMY OF SCIENCES [Proc. 4tii Ser. 





O 



@ 




o 




_J 




Q_ 




^" J 


E 


1 


E 


1 


O 



o 

Q_' 



Vol. XXXVIII] COMPAGNO: SYSTEM ATICS OF GENUS HEMITRIAKIS 71 

parallel to the jaw axis and its vertical sides perpendicular to it. Teeth in 
upright functional positions at the edge of the jaw have outer and inner faces 
on their planes of compression. The orientations of these faces are reversed when 
the teeth are in replacement position but as a convention the functional orienta- 
tion is used here for any tooth. 

The root has its inner face partially formed into a flattened attachment sur- 
face that seats in the dental membrane against the jaw surface. The root has a 
vertical transverse groove that superficially divides the attachment surface into 
two lobes and may extend over the extreme rim of the root to form a transverse 
notch. The outer face of the root may have a strong basal groove extending 
horizontally across it that is overlapped by a strong basal ledge of the crown 
foot. A series of vertical transverse ridges may be present on the basal ledge and 
often extend distally on the outer face of the crown. In many species of Mustelus 
the crown inner face has a rounded protuberance or peg. The peg of one tooth 
extends into the basal groove of the next tooth in succession in the same row, an 
arrangement that may serve to interlock the teeth in the pavement dentitions of 
these forms. 

The distal part of the crown, as opposed to the foot, may have its margin in 
the plane of its compression formed into a sharp cutting edge, with or without 
serrations. Pointed projections from the crown edge are termed cusps or cusplets 
according to their size relative to each other. In carcharhinids a median primary 
cusp is commonly present and is usually larger than other projections of the 
crown edge (when such are present). The primary cusp may have its axis per- 
pendicular or oblique to the tooth base. Its proximal origin may occupy all or 
only part of the foot. When a primary cusp origin is restricted, the adjacent 
crown edges may be formed into other cusps or cusplets, sharp-edged blades, or 
rounded shoulders. 

The planes of compression in the teeth of carcharhinid sharks have their hori- 
zontal sides parallel to the jaw axis, but this axis changes from nearly perpendic- 
ular to the body axis at the symphysis to nearly parallel with the body axis at 
either end of the dental arcade. The horizontal sides of the planes of compression 
for tooth roots and crowns are therefore oriented in an anteromedial-to-postero- 
lateral direction relative to the anterior-to-posterior horizontal body axis along 



FiGURE 4. Outer views of carcharhinid teeth. All teeth except C from left half of 
dental band. A, upper Sth tooth, and D, lower 10th tooth, of Galeorhinus zyoptenis 
(LJVC-0114; 1670 mm. male). B, upper 10th tooth, and E, lower 9th tooth, of Hypogaleus 
sanzibariensis (1220 mm. male; modified from Smith, 1957b). C. Upper tooth of third 
row from end of dental band, Proscyllium habereri (GVF-Hong Kong-88; 523 mm. female). 
F. Same of Triakis semifasciata (LJVC-0137; 1097 mm. male). Abbreviations as in 
Figure 3, except for: PMC, premedial cusplets. 



72 



CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 



® 




© 




All scale lines 1.0 Mm. 



DCL 




DCL 



^NA 



Figure S. Transverse views of carcharhinid vertebral calcification patterns, with 
calcified areas indicated in black. A. Eridacnis barbouri ("Silver Bay" 3514; 258 mm. 
female). B. Proscyllium habereri (UMMZ-179065 ; 565 mm. male). C. Mustelus henlei 
(LJVC-0020; 630 mm. female). D. Hemitriakis japanka (SU-12667). Abbreviations: 
DCL, diagonal calcified lamellae; I, intermedialia; NA, neural arch; NC, notochordal 
canal. 



most of the jaw, with the angle between the sides and the body axis decreasing 
from symphysis to rictus. It is possible with these orientations to distinguish 
anteromedial and posterolateral edges on the crowns and anteromedial and pos- 
terolateral lobes on the roots of most teeth. Exceptions occur at the symphysis, 
where teeth may have medial-to-lateral orientation, and at the ends of the dental 
arcade, where teeth can have anterior-to-posterior orientation. As a convention 
the anteromedial-to-posterolateral relations are used for all teeth. For brevity, 
structures having an anteromedial orientation on the tooth are termed premedial, 
whereas posterolateral structures are postlateral. Thus, carcharhinid teeth can 
have premedial and postlateral cusp edges, cusplets, serrations, blades, etc. 
The terms row and series were used almost interchangeably by Bigelow and 



Vol. XXXVIII] COMPAGNO: SYSTEMATICS OF GENUS HEMITRIAKIS 73 

Schroeder (1948), but Applegate's (1965) usage is followed here. A row is a 
single replicating file of teeth approximately transverse to the jaw axis that in- 
cludes both functional teeth and their replacements in various stages of develop- 
ment. The row represents an entire family of teeth derived from one germinal 
area on the jaw. The term "series" is used for a line of teeth along the jaws 
which is parallel to the jaw axis and includes teeth from all rows present. In 
some carcharhinids, especially those with pavement dentitions and very numerous 
teeth, the concept of series may be meaningless as all teeth are closely adpressed 
in quincunx formation and do not form distinct transverse lines. 

As indicated by Applegate ( 1965 ) , there are two primary types of heterodonty, 
or differentiation between teeth in various positions on the jaws, that can be 
demonstrated in sharks. The first, here termed dignathic heterodonty, involves 
differences in morphology between teeth in opposition or approximate opposition 
in the upper and lower jaws. Dignathic heterodonty can apply to all opposing 
teeth in both jaws or to only some of them. The second type, monognathic 
heterodonty, involves differences between teeth in different positions on the same 
jaw series. IMonognathic heterodonty is not restricted to situations in which ad- 
jacent teeth differ strongly in morphology, but also applies when a tooth in one 
position is different from that in another position on the same series but has a 
gradient of intermediate teeth between itself and the second tooth. The first 
condition can be called disjunct monognathic heterodonty; the second, gradient 
monognathic heterodonty. 

Applegate (1965) used a row-group terminology for implied disjunct monogna- 
thic heterodonty in the dentitions of Odontaspis taurus (Odontaspidae) and 
other sharks. The terms symphysials, alternates, and medials were used for dif- 
ferent tooth types in the region of the symphysis. Remaining teeth were grouped 
into anteriors, intermediates, laterals, and posteriors from premedial to postlateral 
along the dental band. Analogs of the intermediates in lamnoids do not exist in 
carcharhinids. However, some carcharhinid genera, especially those in the ad- 
vanced and intermediate groupings mentioned below, show strong disjunct monog- 
nathic heterodonty and have medials, alternates, symphysials, anteriors, laterals, 
and posteriors. Other genera (as Hemitriakis) have disjunct variation only be- 
tween the medials or alternates at the symphysis and the adjacent parasym- 
physial rows, which may be termed anteroposteriors. 

Two additional types of heterodonty can be defined here. Ontogenic hetero- 
donty is a gradient phenomenon in which tooth morphology at a functional series 
position in a single row or many rows changes with replacement of teeth during 
growth. Gynandric heterodonty, or dental sexual dimorphism, includes differ- 
ences in morphology of teeth in approximately similar series and row positions be- 
tween two individuals or groups of individuals of opposite sex and same species 
at about the same developmental stage. 



74 



CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 




FRT 



UND 





TM 
STM 

STN 



HL 
UPVM 
LPVM 



PRVM 






Figure 6. Carcharhinid fin terminology. A. Idealized first dorsal fin. B. Caudal fin. 
Abbreviations: AM, anterior margin; AP, apex; BA, base; DM, dorsal margin; EL, 
epural lobe; FRT, free rear tip; HL, hypural lobe; IM, inner margin; IN, insertion; 
LOR, lower origin; LPVM, lower postventral margin; OR, origin; PM, posterior margin; 



Vol. XXXVIII] COMPAGXO: SYSTEMATICS OF GEM'S HEMITRIAKIS 75 

Vertebrae. Terminology for vertebral calcified parts follows Ridewood 
(1921). 

Springer and Garrick (1964) subdivided the vertebral complements of sharks 
into precaudal and caudal centra. They noted that an alternative grouping into 
monospondylic and diplospondylic centra was possible, but this was not utilized 
in their study. 

For present purposes the Springer and Garrick dichotomy of vertebral types 
is modified into a three-group system of monospondylic precaudal (MP), diplo- 
spondylic precaudal (DP), and diplospondylic caudal (DC) centra. This 
trichotomy is applicable to most carcharhinids, but breaks down in sharks such 
as Galeorh'inus zyoptcrus where alternating long and short centra of a "stutter 
zone" mark the transition from MP to DP centra. Springer and Garrick's 
method of delimiting the caudal centra at the upper precaudal pit or upper 
caudal origin is followed here despite its shortcomings. 

In some instances it is useful to compare relative numbers of centra in dif- 
ferent vertebral groups of sharks with differing total vertebral counts. A system 
used here divides the IMP, DP, and DC counts by the MP count to give DP/MP 
and DC/ IMP ratios that vary sufficiently between carcharhinid genera and 
species to be of systematic value (MP/MP = 1.00). An alternate system is to 
divide MP, DP, and DC counts by total count and multiply by 100 to obtain 
percent total count for each vertebral group. 

Fins. The terminology used here for carcharhinid fins is explained by fig. 6. 
The following terms apply to paired and unpaired fins other than the caudal: 
Origin; anterior margin; apex; posterior margin; free rear tip; inner margin; 
insertion; and base. The caudal fin terminology includes: Hypural lobe; epural 
lobe; terminal sector; subterminal notch; ventral lobe; dorsal margin; terminal 
margin; subterminal margin; upper postventral margin; lower postventral mar- 
gin; preventral margin; upper origin; and lower origin. 

Genus Hemitriakis Herre, 1923 

Type species. Hemitriakis Icucoperiptera Herre, 1923, by original designa- 
tion. 

Species. There are two named species: H. leucoperiptera, from the Philip- 
pine Islands (detailed distribution in Herre, 1953); and H. japanica (Miiller 
and Henle), from Japan, Taiwan, and Amoy (Chen, 1963). 

Hemitriakis japanica was placed in the genus Galeorhinus Blainville (or its 
junior synonyms, Galeus Cuvier, 1817, not Rafinesque, 1810, and Eugaleus Gill) 



PRVM, preventral margin; STM, subterminal margin; STN, subterminal notch; TM, 
terminal margin; TS, terminal sector; UND, undulations in dorsal caudal margin; UOR, 
upper origin ; UPVM, upper postventral margin ; VL, ventral lobe. 



76 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

by all previous writers (summarized in Fowler, 1941, and Chen, 1963). The 
species was named Galeus japaniciis by IMiiller and Henle (1841). 

Specimens of Galeorhinus japanicus were compared with specimens of Gale- 
orhinus australis, G. chilensis, G. galeus (type species of Galeorhinus), G. 
zyopierus, and with specimens and Herre's (1923) description of Hemitriakis 
leucoperiptera. This indicated that ''japanicus'' does not belong to Galeorhinus 
but is congeneric with Hemitriakis leucoperiptera. 

The two Hemitriakis species are close but H. leucoperiptera differs from H. 
japanica in having the first dorsal origin over inner pectoral margin {H. japanica 
with origin posterior to free rear tip of pectoral). The distance from pectoral free 
rear tips to pelvic origins about equal to first dorsal length from origin to free 
rear tip in H. leucoperiptera but much greater in H. japanica. Hemitriakis leu- 
coperiptera also has fewer vertebrae, with about 144-146 total count (2 speci- 
mens) and 34-35 MP centra {H. japanica with 156-161 total and 41-43 MP 
centra for 7 specimens ) . 

An undescribed Hemitriakis species may be represented by 4 specimens (SU- 
40097) that differ from the sympatric H. leucoperiptera in various proportions, 
fin shapes, and in their strikingly barred and spotted coloration (H. leucoperip- 
tera and //. japanica have a nearly plain coloration). 

Redefinition and Description of the Genus Hemitriakis. Head flat- 
tened dorsoventrally, its length from snout tip to 5th gill opening about % of 
total length. Eyes high on sides of head, above horizontal head rim and level of 
nostrils by a space equal or greater than eye height (fig. 7A). Strong subocular 
ridge present, in dorsal view separating eyes from horizontal head rim by a wide 
space (fig. lA). Eyes not visible in ventral view of head (fig. ID). Eyes elon- 
gate, their apertures over twice as long as high, with a notch present posteriorly in 
adults and subadults. NLE external (fig. 7A), with its edge horizontal. Edge of 
SLE strongly differentiated. Subocular pouch shallow but well defined, with its 
interior surface covered with denticles. 

Spiracles present, slitlike or porelike, 1/5 to 1/7 of eye length. External gill 
openings short, the longest (3rd) less than eye length. Gill rakers absent from 
internal gill openings. 

Nostrils narrow, far apart, a nostril width about 2V2 times in internarial width. 
Anterior nasal flap a short rounded lobe, not a pointed barbel. Nostrils about 
half as far from mouth as from snout tip. Nasoral grooves absent. 

Mouth crescentic, broad, at least 2% times as wide as long. Large papillae 
absent from buccal cavity. Moderately long labial furrows present, upper about 
IV2 times as long as nostril width, the lower % to % of upper. Upper labial fur- 
rows extending anteriorly to below first U of eye. 

Dignathic heterodonty weak, with upper anteroposterior teeth slightly larger 
and with higher crowns and more erect cusps than lowers ; upper medials smaller 



Vol. XXXVIII] COMPAGXO: SYSTEMATICS OF GEXUS HEMITRIAKIS 



77 



® 






Figure 7. Lateral view of carcharhinid heads. 
B. Galeorhinu.s zyoptenis (LJVC-0238). 



A. Hemitriakis japanica (SU-12677). 



than lower ones. Disjunct monognathic heterodonty indicated by differentiation 
of 3 to 6 rows of medials in upper and lower jaws. ^Medial teeth differ from an- 
teroposteriors in their lesser size and erect primary cusps, flanked by 1 or 2 



78 C ALIFORM A ACADEMY OF SCIENCES [Proc. 4th Ser. 

premedial and postlateral cusplets. The sharp-edged anteroposteriors are larger, 
compressed, bladehke cutting teeth with a strong obhque primary cusp and no 
premedial cusplets (fig. 3). Anteroposteriors show strong gradient monogna- 
thic heterodonty. In more premedial teeth in adults and subadults the primary 
cusp is large and has 2 to 4 postlateral cusplets flanking it. From premedial to 
postlateral the crowns of teeth become lower, the primary cusps become more 
oblique, and the postlateral cusplets become fewer and finally disappear. The 
most postlateral anteroposteriors have primary cusps reduced or absent and are 
very low and sharp-keeled. Ontogenic heterodonty present in more premedial 
teeth of anteroposteriors. These teeth gain more postlateral cusplets with growth, 
so that late embryos have no cusplets and adults have 2 to 4 cusplets. Gynandric 
heterodonty not apparent. 

Teeth moderately large, base width of longest lower anteroposteriors about 
0.356-0.405 percent of total length in H. japanica. Tooth rows relatively few; 
Chen (1963) gives 23-29/27-33 (4 specimens) and Tang (1934) gives 35/33 
(1 adult male) total tooth row counts for H. japanica. The 7 examples of H, 
japanica studied here have 33-38/29-33 rows. Herre (1923) gives 18/34 rows 
for the holotype (adult female) of H. leiicoperiptera, but this may be erroneous 
as 33/30 rows were counted in one of the SU-27118 specimens (late embryo). 
One to 5 series functional along jaw edges. Teeth of adjacent rows in the alter- 
nate overlap pattern of Strasburg (1963). Serrations absent from crown edges. 
Crown premedial edge not indented and differentiated. Crown foot with a 
strong basal ledge overlapping a deep basal groove. Transverse ridges present 
on basal ledge only, not extending onto primary cusp. Roots low, deep, with 
strong transverse groove dividing attachment surface into 2 lobes and extending 
through extreme rim of root to form strong transverse notch. Teeth not notice- 
ably protruding when mouth is closed. 

Trunk not compressed, subcylindrical. Interdorsal ridge present. Lateral 
dermal keels absent from caudal peduncle. No precaudal pits. Head-trunk 
length from snout tip to cloaca equal to, or somewhat shorter than, tail length 
from cloaca to caudal tip. 

Denticles from sides of body below first dorsal fin small, with crowns much 
longer than wide at all sizes. A single strong medial cusp and bifurcated lon- 
gitudinal ridge with a weak lateral ridge on each side of crowns of adult denticles. 
In late embryos to subadults medial ridge not bifurcated and lateral ridges 
absent. A pair of weak lateral cusps often present on denticles, but these are 
not constant. 

Pectorals moderately large, pectoral area slightly greater than first dorsal 
area. Pectoral anterior margin about 1% times as long as combined base and 
inner margin lengths. Apex of adpressed pectoral slightly posterior to its free 
rear tip when pectoral inner margin is held parallel to body axis. Origin of 



Vol. XXXVIII] COMPAGNO: SYSTEMATICS OF GENUS HEMITRIAKIS 79 

pectoral below or slightly anterior to fourth gill opening. Pectoral skeleton 
projecting about -i to V2 of pectoral anterior margin length into the fin. Distal 
pectoral radials slightly longer than proximal ones, with broad, truncate tips. 

Pelvics relatively small, their anterior margins less than V2 the length of 
pectoral anterior margins. Pelvic bases closer to 1st dorsal base than to 2nd 
dorsal base. 

Midpoint of 1st dorsal base almost equidistant between pelvic and pectoral 
bases or definitely closer to pectoral bases. First dorsal free rear tip anterior 
to pelvic origins. 

Second dorsal nearly as large as first one, with its height 70 to 80 percent 
of 1st dorsal height. Posterior margin of 2nd dorsal strongly concave. 

Anal much smaller than 2nd dorsal, its height % that of 2nd dorsal and its 
base only % to % of 2nd dorsal base. Anal posterior margin strongly concave in 
adults, shallowly concave in late embryos. Anal origin posterior to 2nd dorsal 
origin by about Va of the 2nd dorsal base length. Anal insertion varying from 
under 2nd dorsal insertion to much less than ^t of 2nd dorsal base length posterior 
to it. 

Caudal with preventral and postventral margins expanded as a short ventral 
lobe in adults and subadults, but scarcely developed in late embryos. Preventral 
caudal margin over % of dorsal caudal margin in adults and subadults, slightly 
shorter in young. Postventral margin differentiated into upper and lower parts 
in subadults and adults, with upper postventral margin % to % of dorsal caudal 
margin. Subterminal caudal margin long, over V2 of terminal caudal margin 
length. Caudal short, dorsal margin about equal to head length and less than % 
of total length. No lateral undulations in dorsal caudal margin. Terminal sector 
of caudal short; distance from subterminal notch to caudal tip only about 2% to 
3 times in dorsal caudal margin. Vertebral axis of caudal slightly raised above 
body axis. 

Vertebrae moderately numerous, total count 144-161. Separation between 
MP and DP centra not sharp, gradual along two transitional centra. Vertebral 
calcification pattern of Applegate's (1967) "carcharhinoid" type. 

Chondrocranium very similar to that of Furgaleus ventraUs as illustrated by 
Whitley (1948) and to that of Mustelus species described and illustrated by 
Gegenbaur (1872) and Holmgren (1941). Supraorbital crest of cranium strongly 
developed and entire. 

Intestinal valve of spiral type, with 6 to 8 turns in the spiral. 

Hemitriakis is livebearing and probably viviparous. Yolk-sac placentae are 
present on the SU-40097 (late embryo) specimens. 

FAMILIAL CLASSIFICATION OF HEMITRIAKIS 
The familial classification of the genus Hemitriakis is troublesome because 
one of its species, H. japanica, is conventionally placed in the family Carcharhin- 



80 C ALIFORM A ACADEMY OF SCIENCES [Proc. 4x11 Ser. 

idae as delimited by Bigelow and Schroeder (1948). In contrast, H. lencoper- 
iptera is usually placed in the family Triakidae. Hemitriakis cannot simulta- 
neously reside in both families, but the problem goes beyond deciding in which 
family this genus belongs. This is because Hemitriakis is almost exactly inter- 
mediate between the Triakidae and Carcharhinidae as defined by modern writers. 
Hence the selection of a family for Hemitriakis is dependent on the validity of 
separating the Triakidae from the Carcharhinidae. 

According to Bigelow and Schroeder (1948) and to Garrick and Schultz 
(1963), triakids differ from carcharhinids only by NLE morphology and denti- 
tion. The triakids are supposed to have rudimentary, external, and transitional 
NLE types (except for Leptocharias and Triacnodon with an internal NLE), 
whereas carcharhinids have an internal NLE. The teeth of triakids are small, 
crushing molariform or bladelike multicuspidate types that are present in several 
functional series on the jaw sides. Carcharhinid teeth are small to large, blade- 
like, with not more than 1 or 2 series of teeth functional at the sides of the jaws. 

As noted by Garrick and Schultz, the separation of the two families is con- 
founded by the seemingly intermediate positions of Triacnodon, Leptocharias, 
and Hemitriakis japanica. Triacnodon especially strains the classification by 
having "triakid" teeth and an internal NLE. However, Gohar and Mazhar 
(1964) claimed that Triacnodon belonged in the Carcharhinidae because it has 
a scroll intestinal valve as in Carcharhinus and other advanced genera. An un- 
published study of the morphology of Triacnodon obcsus confirms Gohar and 
Mazhar's results on the valvular intestine and also demonstrates that Triacnodon 
is very different from other "triakids" in its cranial morphology, pectoral fin 
skeletal structure, head morphology, and many other characters. Of the various 
triakid and carcharhinid genera, Negaprion is evidently closest to Triacnodon. 
The teeth of Triacnodon superficially resemble those of other "triakids" only 
in having premedial and postlateral cusplets flanking a primary cusp, but are 
otherwise strikingly different in the advanced morphology of their crowns and 
roots. It is probable that the "triakid" characters of Triacnodon are convergent 
ones. 

Even without Triacnodon to complicate the issue, the familial separation of 
Triakidae from Carcharhinidae, using the traditional characters, fails when 
other genera are considered. Thus, Hemitriakis had bladelike, sharp-edged 
anteroposterior teeth in 1 to 5 functional series that closely resemble those of 
Galeorhinus, but has medials with multiple premedial and postlateral cusplets 
closely resembling "triakid" teeth. Its NLE is external, as in many, but not all, 
supposed triakids. Furgaleus combines Galcorhiniis-Vike upper anteroposterior 
teeth, Hemigaletis-Yike lower anteroposteriors with one erect primary cusp 
and no cusplets, and an external NLE. Furgaleus is conventionally placed 
in the Triakidae. Another triakid, Leptocharias, has an internal NLE and 



Vol. XXXVIII] COMPAGXO: SYSTEMATICS OF GENUS HEMITRIAKIS 81 

anterolateral teeth with primary cusps, premedial cusplets, and postlateral 
cusplets in the "triakid" pattern. In the genus Galcorhinus (a presumed 
carcharhinid ) , young specimens of G. australis, G. chilensis, G. galeus, and 
G. zyopterns have a transitional NLE, but half-grown to adult individuals 
have these structures internal. Finally, Triakis semijasciata and T. scylUa have 
an external NLE in young specimens but this changes to a transitional or fully 
internal one in adults and subadults. Adult and subadult Mustelus commonly 
have a transitional NLE, but large M. canis may have the internal type (Garman, 
1913; Bigelow and Schroeder, 1948). The teeth of Triakis semijasciata are ar- 
ranged in only 3-4 functional series on the jaw edge. Also, in T. semijasciata the 
teeth show considerable ontogenic heterodonty, with loss of premedial and postlat- 
eral cusplets as the dentition is replaced until many to almost all of the teeth in 
adult specimens have only a strong, oblique primary cusp. Triakis maculata also 
shows a similar type of ontogenic heterodonty (Kato, Springer, and Wagner, 
1967). 

The orthodox distinction of Triakidae from Carcharhinidae is untenable at 
present because the supposedly diagnostic and traditional characters used to 
separate these families fail to do so. As the above examples show, there are 
enough transitional genera and species to make the retention of the two families 
Triakidae and Carcharhinidae an arbitrary choice based on tradition and con- 
venience. I prefer to submerge the Triakidae in the Carcharhinidae. This has 
the obvious disadvantage of creating a huge, unwieldy, and heterogeneous com- 
plex that combines advanced forms with scyliorhinoid genera. However, it may 
be possible eventually to divide the family Carcharhinidae as here constituted 
into a number of lesser families using new characters of comparative morphology 
that are now being investigated. 

Hence, the genus Hemitriakis is considered a member of the expanded family 
Carcharhinidae. 

COMPARISON WITH OTHER GENERA 

This section demonstrates the distinctness of Hemitriakis from other car- 
charhinid genera. A series of synoptic keys is presented in which allied groups of 
genera are compared and contrasted with Hemitriakis. A general key to car- 
charhinid genera is not offered here as revisional studies on the family are in- 
complete at present. 

To facilitate comparison of Hemitriakis with certain genera, it was necessary 
to include some species rearrangements within them in the following discussions. 
This primarily involved removal of some species from the heterogeneous genera 
Triakis and Galeorhinus and proposal of tentative new generic arrangements to 
accommodate them. 

Advanced and Intermediate Carcharhinids. A large proportion of car- 



82 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

charhinid genera comprise the two groups here termed the advanced and inter- 
mediate carcharhinids. The advanced genera inchide Aprionodon, Car char hinus, 
Galeocerdo, Hypoprion, Isogomphodon, Lamiopsis, Loxodon, Negaprion, Prio- 
nace, Rhizoprionodon, ScoUodon, and Triaenodon. The intermediate genera are 
Hemigaleus (including Chacnogaleus, Negogaleus, and Paragaleus) and Hemi- 
pristis (including Dirrhizodon and Heterogaleus). 

The advanced carcharhinids are so named because they depart furthest of 
all genera in the family from the morphology of generalized scyliorhinid genera 
widely thought to occupy the most primitive position among carcharhinoid 
sharks. The Sphyrnidae (hammerheads) is closely allied to the advanced car- 
charhinids but is not included for comparison with Hemitriakis because of its 
unique and obvious specializations. The intermediate genera are very similar to 
the advanced ones in many characters, but retain some generalized features in 
the morphology of the cranium, fins, dentition, and intestinal valve. 

The advanced and intermediate carcharhinids are grouped together for 
brevity to compare them with Hemitriakis. The following synopsis covers only 
a representative sample of numerous differences between these genera and Hemi- 
triakis. 

la. Eyes low on head, their ventral margins meeting or extending across the horizontal 
head rim. Subocular ridge weak or obsolete. NLE always internal, with slanted edge. Sub- 
ocular pouch very deep, with inner surface of SLE and bottom of pouch lacking denticles. 
Crowns of teeth without a strong basal ledge and groove (except in lower teeth of Hemi- 
galeus) . Transverse ridges virtually absent from crown foot. Denticles of adults as wide or 
wider than long, with three or more subequal cusps and ridges. Precaudal pits present always 
at upper caudal origin and usually at lower origin also. Pectoral skeleton projecting at 
least % of pectoral anterior margin length into fin, with distal radials much longer than 
proximals. Distal radials with tapering, acute tips. .Anal large relative to 2nd dorsal, its 
height 70 per cent or more of 2nd dorsal height. Lateral undulations present along dorsal 
caudal margin (except in young of some species and in ScoUodon where undulations are in- 
differently developed). Intestinal valve a scroll in advanced carcharhinids, but a spiral with 
only 2-6 turns in Hemigaleus and Hcmipristis. Chondrocranium with isolated preorbital and 
postorbital processes only, without an intermediate supraorbital crest . . . Advanced and 

Intermediate Carcharhinids. 

lb. Eyes high on head, their ventral margins widely separated from the horizontal 
head rim in ventral view. Subocular ridge very strong. NLE external, with edge 
horizontal. Subocular pouch shallow, with denticles covering its internal surface. 
Crowns of teeth with strong basal groove and ledge. Transverse ridges irregularly present 
on crown foot. Denticles of adults longer than wide, with a very strong medial ridge and 
cusp and flanking weak lateral ridges also ; weak lateral cusps irregularly present. Precaudal 
pits absent. Pectoral fin skeleton projecting only % to V> of pectoral anterior margin dis- 
tance into fin, with distal radials slightly longer than proximals. Distal radials with parallel 
articulating edges and truncate tips. Anal relatively small in relation to second dorsal, its 
height only Mj that of second dorsal height. Lateral undulations of dorsal caudal margin 
absent. Intestinal valve a spiral, with 6-8 turns. Chondrocranium with strong supraorbital 
crest between preorbital and postorbital processes Hemitriakis. 



Vol. XXXVIII] COAIPAGXO: SYSTEMATICS OF GEXUS HEMITRIAKIS 83 

Galeorhinus and Allied Genera. Recent workers have included the fol- 
lowing species in Galeorhinus: G. galeus (Linnaeus, 1758) ; G. japanicus (Miiller 
and Henle, 1841); G. austraUs (Macleay, 1881); G. zyopterus (Jordan and 
Gilbert, 1883) ; G. chilensis (Perez Canto, 1886), including G. molinae (Philippi, 
1887); G. omancnsis (Norman, 1939); G. hyugaensis Miyosi, 1939; G. vita- 
minicus de Buen, 1950; and G. zanzibariensis Smith, 1957. Of these nine species, 
four are sufficiently different to require removal from Galeorhinus. "Galeorhinus" 
japanicus has been already transferred to Hemitriakis. G. hyugaensis and the 
closely similar G. zanzibariensis are placed in the genus Hypo galeus and discussed 
below. ^'Galeorhinus" omanensis, as suggested by its describer (Norman, 1939), 
is not congeneric with Galeorhinus and will be discussed in a forthcoming paper 
by Mr. Stewart Springer and myself. It is included in the synopsis below to dis- 
tinguish it from Hemitriakis. 

The remaining 5 nominal species comprise the genus Galeorhinus as here 
delimited. Garman (1913), Fowler (1929, 1941), and Bigelow and Schroeder 
(1948) considered G. zyopterus a junior synonym of G. galeus, while Kato, 
Springer, and Wagner (1967) tentatively synonymized G. chilensis with G. 
zyopterus. McCoy (1885) compared G. australis with G. galeus and listed 
several proportional differences between a few specimens of G. australis 
and one of G. galeus. However, comparison of a pair of equal sized specimens of 
G. galeus and G. australis suggests that most, if not all, of McCoy's differences 
were allometric ones based on comparison of dissimilar sized specimens. G. vitam- 
inicus, as described by De Buen (1950), is hardly different from other Galeo- 
rhinus species. It may be that all 5 species are synonyms, as Smith (1957b) 
maintained, but the validity of this hypothesis cannot be tested at present because 
of insufficient material. 

Smith (1957b) proposed the subgenus Hypogaleus for its type, Galeorhinus 
(Hypogaleus) zanzibariensis Smith, 1957, and for Hemitriakis japanica. Accord- 
ing to Smith, Hypogaleus species have teeth without the transverse notch on 
their roots, but in Galeorhinus (including only G. galeus) the notch is present. 
In Galeorhinus the caudal terminal sector is large, about Vi caudal length, but 
much smaller and less than V^ caudal length in Hypogaleus. Hypogaleus has the 
second dorsal at least twice as great in area as anal, but Galeorhinus has these 
fins subequal in size. In Galeorhinus the pelvic fins of adults are inserted 
behind the middle of the total length; in Hypogaleus the pelvics of adults are 
inserted well in advance of the middle of the total length. 

Apparently Smith used only literature descriptions for Hemitriakis japanica. 
Although most of the fin characters for Hypogaleus fit Hemitriakis japanica, the 
dentitional character does not, as this species has a strongly developed transverse 
notch. Also, Smith's tooth photographs of G. {Hypogaleus) zanzibariensis seem 
to indicate that this species has much higher roots and obsolete basal ledges and 



84 



CALIFORNIA ACADEMY OF SCIENCES 



[Proc. 4th Ser. 




Figure 8. Lateral views of carcharhinid sharks. A. Hemitriakis japanica (SU-12677). 
B. Hypogaleus zanzibariensis (modified from Smith, 19S7b). C. Galeorhinus zyoptcrus 
(LJVC-0238). 



Vol. XXXVIII] COMPAGXO: SYSTEMATICS OF GENUS HEMITRIAKIS 85 

grooves on its teeth (figs. 4B, 4E). Galeorhinus and Hemitriakis as presently 
delimited have low roots and strong basal ledges and grooves (figs. 3, 4A, 4D). 

Also, Smith did not examine Miyosi's (1939) description of Galeorhinus 
hyitgaensis. Comparison of the accounts of G. hyugaensis by Miyosi and by 
Chen ( 1963) with those of G. zanzibaricnsis by Smith (1957b) and by D'Aubrey 
(1964) indicates that these species are virtually identical in all important details 
of morphology (including dentition) and coloration. Indeed, it will be necessary 
to compare specimens of the two species to determine what differences, if any, 
exist between them. 

Galeorhinus zanzibaricnsis and G. hyugaensis are close to Galeorhinus 
proper and to Hemitriakis but are sufficiently different to merit generic status. 
Hence I propose to raise Hypogaleus Smith from subgenus to genus and include 
in it the two nominal species H. hyugaensis (Miyosi, 1939) and H. zanzibarien- 
sis (Smith, 1957). 

The Australian genus Fwgalcus is included here because its two species, as 
described and illustrated by Whitley (1943a, 1943b, 1944, 1948), have upper 
anteroposterior teeth that strongly resemble those of Hemitriakis^ Hypogaleus, 
and Galeorhinus. Furgaleus is closest to Hemitriakis but is easily distinguished. 

The following is a synopsis of Galeorhinus and allied genera (including 
Hemitriakis). 

la. Postlateral cusplets absent from anteroposterior teeth in upper jaw. Upper medial 
teeth without cusplets. First dorsal fin far forward, with origin over anterior half of 
pectoral base. Caudal fin without ventral lobe in adults "Galeorhinus^^ omanensis. 

lb. Postlateral cusplets present on upper anteroposterior teeth. Upper medials with 
both premedial and postlateral cusplets. First dorsal origin posterior to pectoral base inser- 
tion. Caudal with moderate to strong ventral lobe in adults 2. 

2a(lb.). Nostrils larger and closer together, their widths about twice in internarial 
width. Nostrils equidistant between snout tip and mouth. Anterior nasal flap formed into 
a long, slender barbel. Dignathic heterodonty strong, with upper anteroposteriors having 
oblique primary cusps and postlateral cusplets while lowers have erect primary cusps and 
no cusplets Furgaleus. 

2b(lb.). Nostrils smaller and farther apart, their width 2^2 times in internarial width 
or more. Nostrils much closer to mouth than to snout tip. Anterior nasal flap not produced 
into a barbel. Dignathic heterodonty weak ; upper and lower anteroposteriors with oblique 
primary cusps and postlateral cusplets 3. 

3a(2b.). Eyes high on sides of head, above level of nostrils by a space equal to 
or greater than eye height. Eyes over twice as long as high. NLE external in adults and sub- 
adults, with horizontal edge. Anterior nasal flap moderately large, expanded as a rounded 
lobe. Posteriormost anteroposterior teeth elongate, carinate. Interdorsal ridge present. 
Adult and subadult denticles with crowns much longer than wide and with lateral cusps 
and ridges weak. Caudal with short ventral lobe in adults and subadults (fig. 8A) 
Hemitriakis. 

3b(2b.). Eyes lower on sides of head, above level of nostrils by a space less than eye 
height. Eyes twice as long as high or less. NLE internal in adults and subadults, with diagonal 
edge. Anterior nasal flap reduced, expanded as a minute, pointed lobe. Posteriormost 



86 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

anteroposterior teeth not elongate, carinate. Interdorsal ridge absent. Denticles of adults 
and subadults with crowns nearly as long as wide and with strong lateral cusps and ridges. 
Caudal with long ventral lobe in adults and subadults 4. 

4a(3b.). Head very short, about % of total length in adults (fig. 8B). Subocluar ridge 
strong; in dorsal view eyes separated from horizontal head rim by a moderately wide space. 
Transverse notch absent from tooth roots (fig. 4B). First dorsal as large or larger than 
pectoral. Second dorsal about % as high as first dorsal and about twice as high as anal. 
Terminal sector of caudal about 2.6 in dorsal caudal margin. Upper postventral margin 
nearly ^/4 as long as dorsal caudal margin Hypogaleus. 

4b(3b.). Head longer, over % of total length in adults (fig. 8C). Subocular ridge 
obsolete; in dorsal view ventral eye margins contact horizontal head rim. Transverse notch 
present on tooth roots (fig. 4A). First dorsal much smaller than pectoral. Second dorsal 
less than half as high as first one and subequal to the anal in height. Terminal sector of 
caudal about 2.0 in dorsal caudal margin. Upper postventral margin only about Vi as long 
as dorsal caudal margin. Galeorhinus. 

Leptocharias and Scylliogaleus. Two aberrant, monotypic African 
genera, Leptocharias and Scylliogaleus, differ greatly from Hemitriakis. While 
Scylliogaleus is apparently closest to typical Mustelus species, the taxonomic po- 
sition of Leptocharias is quite isolated in the Carcharhinidae. 

Leptocharias, Hemitriakis, and Scylliogaleus are compared in the following 
synopsis. Additional data on Scylliogaleus is from Boulenger (1902) and Smith 
(19S7c). 

la. Eyes low on sides of head, above level of nostrils by less than eye height. Subocular 
ridge obsolete ; ventral margin of eyes touching horizontal head rim in dorsal view. Eyes 
less than twice as long as high, with a slant-edged, internal NLE (fig. 2D). Spiracles 
minute, porelike, less than %o of eye length. Anterior nasal flap expanded as a pointed 
barbel. Gynandric heterodontia strong, expressed by presence of about 4 tooth rows of 
hypertrophied "anteriors" in both jaws on either side of weakly differentiated medials in 
adult males but not females. Teeth other than anteriors with slender, erect primary cusps 
and both premedial and postlateral cusplets, not bladelike or molariform. Vertebrae very 
numerous, 198-213 total centra (9 examples; data for 2 from Springer and Garrick, 1964). 
Spiral intestinal valve with about 16 turns. Supraorbital crest absent from cranium, with 
isolated preorbital and postorbital processes only Leptocharias. 

lb. Eyes higher, above level of nostrils by an eye height or more. Subocular ridge strong ; 
eyes separated from horizontal head rim by a wide space. Eyes over twice as long as high, 
with horizontal-edged and external NLE (fig. 7A). Spiracles larger, \i>, to Y-; of eye length. 
Anterior nasal flap not formed into a barbel. Gynandric heterodonty not apparent. Teeth 
either molariform or bladelike, without premedial cusplets (except on medials of Hemi- 
triakis). Vertebrae fewer, 143-161 total centra (10 examples). Spiral valve with 6 to 8 
turns. Supraorbital crest present on cranium 2. 

2a(lb.). Snout bluntly rounded, semicircular in shape. Anterior nasal flaps greatly 
enlarged as broad triangular lobes extending posteriorly to overlap mouth. Nostrils very 
large and separated by a distance much shorter than a nostril width. Deep nasoral grooves 
present. Teeth with crowns flattened and rounded to form a crushing pavement as in typical 
Mustelus species. Teeth not differentiated into medials and anteroposteriors. Tooth rows 
60-72 in each jaw; 9-10 series functional in upper jaw, 16-17 in lower (Smith, 1957c). 



Vol. XXXVIII] COMPAGNO: SYSTEMATICS OF GENUS HEMITRIAKIS 87 

Pelvics large, anterior margin lengths M> or more of pectoral anterior margin lengths. Free 

rear tip of first dorsal over or posterior to pelvic origins Scylliogaleus. 

2b(lb.). Snout narrower, parabolic in shape. Anterior nasal flaps small truncate lobes, 
terminating far anterior to mouth. Nostrils smaller, further apart, their widths about 2% 
times in internarial width. Nasoral grooves absent. Teeth differentiated into medials and 
anteroposteriors, not forming a pavement. Anteroposteriors sharp-edged, bladelike teeth, 
with oblique primary cusps and postlateral cusplets ; medials are not bladelike and have 
premedial cusplets also. Tooth rows fewer, 18{ ?)-39/27-34; only 1 to S series functional 
along jaw edges. Pelvics smaller, their anterior margins less than V2 length of anterior 
pectoral margins. Free rear tip of first dorsal anterior to pelvic origins Hemitriakis. 

Triakis and Associated Genera. Included here are those species placed by 
Bigelow and Schroeder (1948) and by Kato (1968) in the genera Triakis, Mus- 
telus, Eridacnis, and Calliscyllium. 

The systematics of Triakis and its relatives is unsatisfactory at present. This 
is in part due to the interpretations of Triakis by Garman (1913), Fowler (1929, 
1941), Bigelow and Schroeder (1944, 1948), Garrick (1954), Kato (1968), 
and Springer (1968), which included several scyliorhinid-like species in this 
genus that are clearly not congeneric with typical Triakis species (as T. scyllia 
and T. semijasciata). Also, the separation of Triakis from Mustelus on differ- 
ences in tooth crown morphology seems untenable, as there are many dentitionally 
intermediate species between "typical" extremes of these genera. Bigelow and 
Schroeder (1940) and Kato (1968) have discussed the latter problem in detail, 
but left the two genera separate. 

Smith (1957a) proposed a solution to the Triakis heterogeneity problem. He 
removed Calliscyllium venustum Tanaka from Triakis and reinstated Calliscyl- 
lium Tanaka as a monotypic genus for it. Also, he proposed the genus Neotriakis 
for his A^. sinuans and for Triakis barbouri Bigelow and Schroeder. Finally, he 
transferred Triakis henlei (Gill) to Mustelus. 

Smith's separation of Calliscyllium and Neotriakis from Triakis is undoubt- 
edly correct, as species included in these scyliorhiniform genera exhibit many dif- 
ferences from typical Triakis. Unfortunately, Smith retained two anomalous 
species, Triakis attenuata Garrick and Hemitriakis leucoperiptera Herre, in the 
genus Triakis. Triakis attenuata is closer to Calliscyllium and Neotriakis in 
the sense of Smith than to Triakis proper, and its presence in Triakis makes 
separation of that genus from Neotriakis especially difficult with the limited and 
ambiguous generic characters utilized by Smith to define these genera. Finally, 
placement of Triakis henlei in Mustelus further undermines the classical tooth 
crown differences purported to separate Triakis from Mustelus; however, T. hen- 
lei is closer to typical forms of Mustelus than to those typical of Triakis in many 
respects. Smith was evidently unaware of the Triakis-Mustelus continuity prob- 
lem, as he later (1957c) gave Mustelus familial separation from Triakis in his 
family Scylliogaleidae (along with Scylliogaleus). 



88 C ALIFORM A ACADEMY OF SCIENCES [Proc. 4th Ser. 

A tentative reclassification of Triakis and associated genera is presented here, 
subdividing these taxa into two groups: 1. Typical forms of Triakis, intergrad- 
ing species, and typical Mustelits forms. 2. Scyliorhiniform triakoids, including 
as subgroups: A. Genus ProscylUum; B. Genus Eridacnis; C. Triakis jehlmanni; 
D. T. attenuata. 

Triakis, Mustelus, and intermediates (or Triakis-Mustclus) are closer to 
Hemitriakis than are other carcharhinids. Triakis-Mustclus includes Triakis scyl- 
lia, T. semijasciata, T. maculata, T. acutipinna, Mustelus henlei, and the various 
other Mustelus species. 

Typical species of Triakis, with strongly cuspidate teeth ( T. scyllia, T. semi- 
jasciata), form one extreme of a dentitional continuum with molariform-toothed 
Mustelus species at the other extreme. The continuum is filled by a host of denti- 
tionally intermediate forms, as T. maculata, T. acutipinna, Mustelus nigropuncta- 
tus, M. henlei, M. dorsalis, M. megalopterus, M. natalensis, and M. higmani, that 
exhibit various stages of cusp and cusplet reduction on tooth crowns. Also, ex- 
amination of small (150-450 mm. total length range) specimens of typical Mus- 
telus species, as M. canis, M. calif ornicus, M. nianazo, M. mustelus, M. lunulatus, 
and M. griseus, indicates that cusps are often well developed on the teeth of small 
individuals and that cusp obsoleteness in larger examples probably results from 
ontogenic heterodonty. 

Although condition of tooth cusps has been the only character regularly 
utilized in separating Triakis from Mustelus, extremes of the former genus differ 
from typical members of the latter by a number of additional characters. These 
include: 1. Absence of peg on inner face of crown and root. 2. Lesser number 
of tooth rows. 3. Lesser number of tooth series. 4. Absence of a tooth pavement. 
5. Bluntly rounded, short snout (versus long, pointed or paraboloid snout in 
Mustelus). 6. Very short, arcuate mouth (versus longer, more angular mouth 
in many Mustelus species). 7. Reproduction ovoviviparous (viviparous in at 
least some Mustelus, including M. henlei). In addition, there are about a dozen 
cranial differences between Triakis semijasciata and 3 species of Mustelus (M. 
henlei, M. calij ornicus, and M. lunulatus, which are virtually identical cranially). 
The brain, cranial nerves, and sense organs of T. semijasciata also differ in 
several respects from those of M. henlei. 

The tooth peg is found in many Mustelus species (including M. henlei), but 
not in Triakis maculata, T. scyllia, or T. semijasciata. Its condition is not con- 
firmed for all species of Triakis-Mustelus and cannot be used to separate the 
two genera at present. Tooth row and series counts apparently vary along a 
continuum as in crown morphology, with an added complication that in at least 
some Mustelus (if not all forms) the tooth row and series counts increase with 
size increase. Tooth pavementization, snout shape, and mouth morphology evi- 
dently show a similar variation spectrum. Data on cranial, neural, and repro- 




Vol. XXXVIII] COMPAGNO: SYSTEMATICS OF GENUS HEMITRIAKIS 89 

ductive characters is not available for many to most Triakis-Mustelus species, 
making it impossible to judge their utility in separating the two genera. 

External morphology suggests that Triakis is not separable from Mustelus, 
but merging the two genera here would be premature with incomplete knowledge 
of promising anatomical characters. However, Triakis-Mustclus is treated as a 
single unit here for comparison with Hemitriakis. 

The scyliorhiniform triakoids include species formerly placed in the genera 
Proscyllium, Calliscyllium, Eridacnis, Neotriakis, and Triakis. They are divisable 
into four subgroups, two of which are provisionally ranked as genera. 

The first genus, Proscyllium, is a structural link between the Carcharhinidae 
and Scyliorhinidae but is placed in the former family because of its anteriorly 
positioned first dorsal fin. 

The systematic treatment of Proscyllium and its synonym, Calliscyllium, by 
various writers has been highly variable and extremely confusing. Hilgendorf 
(1904) proposed Proscyllium as a subgenus of Scyllium Cuvier (= Scylliorhinus 
Blainville), with a single new species, S. {Proscyllium) habereri, from Formosa. 
Later Tanaka (1912) described a new genus and species, Calliscyllium vcnustum, 
from Japan. Tanaka did not mention Hilgendorf's very similar species in his 
account. Although Tanaka considered Calliscyllium a scyliorhinid, Garman 
(1913) placed it in his family Galeorhinidae (= Triakidae) and synonymized 
it with Triakis. Garman also placed Scyllium {Proscyllium) habereri in the 
Catulidae (= Scyliorhinidae) and raised the rank of Proscyllium to genus. 
Schmidt (1930) described and illustrated a Japanese specimen of Proscyllium 
habereri. His account is of special interest as he compared his specimen with 
measurements and photographs of the holotype of Hilgendorf's species and 
found no significant differences between the two specimens. Schmidt's account 
of Proscyllium habereri closely matches Tanaka's description of Calliscyllium 
venustum, but for unknown reasons Schmidt did not refer to Tanaka's account 
or to his own (1928) description of an Okinawan specimen of Calliscyllium 
venustum. White (1937) recognized both Calliscyllium venustum and Proscyl- 
lium habereri as scyliorhinids in a broad sense but placed the former in her 
family Halaeluridae and the latter in her family Catulidae. Fowler (1929, 1941) 
followed Garman in placing Proscyllium in Scyliorhinidae and placed Triakis 
venusta (Tanaka) in the subfamily Triakiinae of the family Eulamiidae or 
Galeorhinidae (= Carcharhinidae). Bigelow and Schroeder (1948) placed 
Proscyllium habereri in the family Triakidae, but did not discuss its 
generic status in that family. These writers followed Carman's synonymy 
of Calliscyllium with Triakis. Garrick (1954) discussed Triakis venusta, 
but not Proscyllium habereri. Smith (1957a) recognized Calliscyllium as 
distinct from Triakis, but also overlooked Proscyllium habereri. Lindberg 
and Legeza (1959) synonymized Proscyllium with Triakis, but considered 



90 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

Triakis habereri distinct from T. venusta. Chen (1963) placed both Proscyllium 
habereri and Triakis venusta in the family Triakidae, but separated Proscyllium 
from Triakis by supposed absence of the NLE in the former genus. Finally, 
Kato (1968) removed CalliscylUum from synonymy of Triakis on reproductive 
differences, but did not mention Proscyllium. 

Comparison of accounts of Proscyllium habereri and CalliscylUum venustum 
with each other and with specimens indicates that Lindberg and Legeza were 
correct in regarding these species as congeneric. However, "venustum'^ and 
"habererP^ are not congeneric with typical species of Triakis and are placed here 
in the genus Proscyllium. CalliscylUum is therefore a junior synonym of Proscyl- 
lium. The two species P. venustum and P. habereri are possibly synonyms also, 
as the small differences between them listed by Lindberg and Legeza (1959) 
may be of only variational and allometric significance. 

The genus Eridacnis includes a few species of deepwater sharklets allied to 
Proscyllium but sufficiently different to merit generic status. 

Eridacnis was established by Smith (1913) for £. radclijjei, a new shark 
from the Philippine Islands. Eridacnis was supposed to differ from Triakis 
Miiller and Henle by lacking labial furrows, but, as Kato (1968) pointed out, 
the holotype of E. radclijjei has vestigial labial furrows presumably overlooked 
by Smith. Bigelow and Schroeder (1944) described as Triakis barbouri a similar 
but specifically distinct shark from Cuba, but did not compare it with Eridacnis 
radclijjei. Misra (1950) described a third form, Proscyllium alcocki, from the 
Andaman Sea. Data from Misra's account indicates that "alcocki^' does not 
belong in Proscyllium as here defined but falls in Eridacnis instead. The species 
"alcocki" closely resembles E. radclijjei and therefore it is quite possible that 
these two names are synonymous (Norman, 1939, reported E. radclijjei from 
the Gulf of Aden, which is west of the type localities of both E. radclijjei and 
'^alcocki'^). Smith (1957a) described a new genus, Neotriakis, for his new 
South African species A^. sinuans. Smith included Triakis barbouri in Neotriakis 
but overlooked Proscyllium alcocki and did not compare Neotriakis species with 
the closely similar Eridacnis radclijjei. Kato (1968) considered the characters 
used to separate Neotriakis and Eridacnis from Triakis to be untenable, and 
synonymized the three genera. However, Kato regarded the species "radclijjei,''^ 
"barbouri,''^ and "sinuans''' as closely related to each other within the genus 
Triakis. Kato's synonymy was adopted unchanged by Springer (1968). 

The genus Eridacnis is revived here for the species E. raclijjei, E. alcocki, 
E. barbouri, and E. sinuans, with Neotriakis considered as a junior synonym. 

Triakis jehlmanni, a small shark recently described by Springer (1968) 
from Somalia, forms a third group closely similar to Proscyllium and Eridacnis 
in many details. Its vertebral calcification pattern and relatively short broad 
caudal are as in Proscyllium, but its vertebral count, vertebral group ratios, short 



Vol. XXXVIII] COMPAGXO: SYSTEM ATICS OF GENUS HEMITRIAKIS 91 

body cavity, nostril spacing, pectoral fin position, first dorsal size, and anal base 
size fit Eridacnis. The blotched and spotted color pattern, extremely short pre- 
caudal tail (distance from cloaca to lower caudal origin about twice in distance 
from snout tip to cloaca), broad head, and stout body distinguish "fehlnianni" 
from both Proscyllium and Eridacnis. Mode of reproduction and clasper mor- 
phology are unknown for the species. Triakis jehlmanni seems closer to Eridacnis 
than Proscyllium but may require subgeneric or generic separation from typical 
Eridacnis species. It does not belong to Triakis-Mustelus as here limited and 
cannot be confused with Hemitriakis. Generic placement of " fehlmannr' is prob- 
lematical at present; therefore the species must be left as a tentative and possibly 
dubious appendage to Eridacnis. 

The New Zealand Triakis attenuata, as described by Garrick (1954), agrees 
with Proscyllium, Eridacnis, and T. jehlmanni in its NLE type, detailed tooth 
morphology, eye position, and large second dorsal, but differs from these forms 
in its elongate snout, narrower and more widely spaced nostrils, longer labial 
furrows, more numerous tooth rows, more anterior position of first dorsal fin, 
origin of second dorsal anterior to anal origin, exceptionally small anal fin with 
base only half length of second dorsal base, weak ventral caudal lobe, and larger 
size. Unfortunately nothing is known of its cranium, pectoral fin skeleton, verte- 
bral calcification pattern, vertebral counts, vertebral group ratios, clasper 
morphology, and buccal cavity. Triakis attenuata presumably forms a tentative 
fourth group of scyliorhiniform triakoids allied to, but distinct from, Proscyl- 
lium-Eridacnis-T . jehlmanni. The species is remote from Hemitriakis and is suf- 
ficiently different from Triakis-Mustelus to be excluded from that group. 
Separate generic status may be required for T. attenuata, but insufficient data 
on the species prohibits a decision on the matter for now. 

The following generic synopsis compares Hemitriakis to Triakis-Mustelus, 
Proscyllium, and Eridacnis (excluding T. jehlmanni). 

la. NLE rudimentary in adults (fig. 2A.). Labial furrows vestigial, confined to corners 
of mouth. All teeth with erect cusps and usually cusplets also (some species have teeth near 
symphysis lacking cusplets). Posterior teeth polycuspidate, combhke in shape (fig. 4C). 
Gradient monognathic heterodonty present, in which premedial cusplets increase in number 
from symphysis to rictus and displace primary cusp from central position on crown foot to 
a postlateral location. Cusps and cusplets do not become obsolete with age. First dorsal fin 
with midpoint of its base closer to pelvic origins than to pectoral insertions. Second dorsal 
origin over or posterior to anal origin. Pectoral fin skeleton as in scyliorhinids, with distal 
radials much shorter than proximal ones. Vertebral centra of thoracic region in adults with 
peripheral calcifications of the intermediaha only, not developed into strong lateral and 
vertical wedges between halves of calcified primary double cone (figs. SA, 5B.). Diagonal 
calcified lamellae of double cone, when present, in form of rounded lobe opposite each 
basidorsal and basiventral. Large papillae present on dorsal and ventral surfaces of buccal 
cavity and pharynx posterior to teeth, forming dermal gill rakers along internal branchial 
apertures 2. 

lb. NLE external, transitional, or internal in adults. Labial furrows well developed. 



92 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

extending far onto jaws. Teeth either cuspless or with cusps that range from erect to 
strongly oblique and often showing monognathic heterodonty in increasing obliqueness 
toward ends of dental band. Posterior teeth carinate, molariform, or weakly monocuspidate, 
not comblike (fig. 4F.). Increase of premedial cusplets and displacement of primary cusp 
postlaterally not apparent in species with cuspidate teeth, but instead cusps and cusplets 
tend to become less prominent postlaterally and may be completely absent on posteriormost 
teeth. Many species (not including Hemitriakis) tend to reduce or lose cusplets or even 
cusps with age. First dorsal fin with base midpoint equidistant between pectoral insertions and 
pelvic origins or closer to pectoral insertions. Second dorsal origin well anterior to anal origin. 
Pectoral fin skeleton with distal radials equal in length to, or longer than, proximal ones. 
Vertebral centra of thoracic region of adults and subadults with intermedialia extending as 
strong lateral and vertical wedgelike calcifications between halves of calcified primary double 
cone (figs. SC, SD.). Diagonal calcified lamellae well developed, extending as thin plates into 
the basidorsals and basiventrals. Papillae absent from buccal cavity, pharynx, and internal 
branchial apertures 3. 

2a(la.). Nostrils very close together, internarial width only V-2 nostril width. Distance 
from hind edge of anterior nasal flaps to mouth only about % of nostril width. Head 
length from snout tip to Sth gill opening shorter than body length from pectoral insertion 
to pelvic origin. First dorsal length from origin to free rear tip less than % length of 
interdorsal space. Anal base length only Mj of distance between anal insertion and 
lower caudal origin. Caudal short, less than % of total length. Greatest height of caudal 
about 1/4 of upper caudal margin. Vertebrae more numerous, total count 146-168 (6 
examples). DP/MP ratios 1.58-1.82; DC/MP 1.08-1.28. Diagonal calcified lamellae of 
trunk centra present as four rounded lobes extending slightly into areas of basidorsals 
and basiventrals (fig. 5B.). Claspers of adult males with a row of recurved clasper hooks 
along external flap of hypopyle. Color pattern of scyUorhinid-Uke spots and stripes 
present. Reproduction oviparous ProscylUum. 

2b(lb.). Nostrils farther apart, internarial width about equal to nostril width. Distance 
from hind edge of anterior nasal flaps to mouth about V-2 of nostril width. Head length 
longer than body length from pectoral to pelvic. First dorsal length % to % of interdorsal 
space. Anal base length subequal to distance between anal insertion and lower caudal 
origin. Caudal longer, over V4 of total length. Greatest height of caudal less than % of 
upper caudal margin. Vertebrae less numerous, total count 113-135 (26 examples). 
DP/MP ratios l.OS-1.45; DC/MP 1.29-1.53. Diagonal calcified lamellae not developed 
in trunk centra (fig. 5A.). Claspers without hooks. Coloration plain or with a few 
obscure stripes confined to tail. Reproduction ovoviviparous as far as is known . _ Eridacnis. 

3a(lb.). Nostrils narrow and farther apart; nostril width about 2V-2 times in internarial 
width. Teeth larger, basal width of largest lower anteroposteriors about 0.356 to 0.405 
percent of total length {H. japanica, 4 examples). Teeth differentiated into medials and 
anteroposteriors. The latter are strongly compressed, bladelike cutting teeth with an 
oblique primary cusp and a few small postlateral cusplets only. Fewer tooth rows present, 
? 18-38/24-34. Pelvic fins with anterior margins less than half as long as pectoral anterior 
margins Hemitriakis. 

3b(lb.). Nostrils wider, closer together; nostril width 2 times or less in internarial. 
Teeth smaller, those of species with largest teeth (Triakis semijasciata) only 0.172 to 0.262 
percent of total length (largest lower teeth, 11 examples) and considerably smaller in other 
species. Teeth not differentiated into distinct medials and anteroposteriors, but showing 
regular gradient monognathic heterodonty between rows in symphysial and parasymphysial 
regions. Teeth corresponding to anteroposteriors of Hemitriakis either cuspless or having 



Vol. XXXVIII] COMPAGNO: SYSTEMATICS OF GENUS HEMITRIAKIS 93 

an erect or oblique median primarx- cusp and usually both premedial and postlateral 
cusplets when cusplets are present. Teeth more weakly compressed, not sharp-edged, 
modified for grasping or crushing. More tooth rows present, 44-80+/33-80+. Pelvic 

fins larger, with anterior margins over Y2 as long as pectoral anterior margins 

Triakis-Mustelus. 

SUMMARY 

The shark Galeus japanicus ]\Iuller and Henle, long considered a species 
of Galeorhinus Blainville (or one of its synonyms), is placed in the genus 
Hemitriakis Herre, which is removed from the synonymy of Triakis ]Muller 
and Henle and redefined. Hemitriakis contains two described species, H. 
japanica (INIiJller and Henle) and H. leucoperiptera Herre. 

The familial position of Hemitriakis is discussed and the separation of 
the families Triakidae and Carcharhinidae is rejected on present evidence. 
Hemitriakis is placed in the expanded family Carcharhinidae. 

Other carcharhinid genera are compared with Hemitriakis in synoptic keys, 
and several tentative systematic rearrangements of species in certain genera 
are presented to facilitate comparison with Hemitriakis. The genus Galeorhinus 
is restricted to the nominal species G. galeus, G. australis, G. zyopterus, G. 
chilensis, and G. vitaminicus, while a former subgenus, Hypogaleus Smith, is 
accorded generic rank. Hypogaleus contains two nominal species, H. zanzi- 
bariensis (Smith) and H. hyugaensis (Miyosi). Consideration of Galeorhinus 
omanensis (Norman) is postponed for another paper. 

In addition to Hemitriakis, four tentative groups of scyliorhiniform triakoids 
are removed from Triakis. The first is the genus Proscyllium Hilgendorf, of 
which Calliscyllium Tanaka is a junior synonym. Proscyllium has two nominal 
species, P. habereri (Hilgendorf) and P. venustum (Tanaka). The genus 
Eridacnis Smith forms the second group, with E. radclijjei Smith, E. barbouri 
(Bigelow and Schroeder), E. simians (Smith), and the dubious E. alcocki 
(]\Iisra) as its constituent species. The last two groups contain Triakis jehl- 
manni Springer and T. attenuata Garrick; these are not given genus-group 
names because of insufficient evidence on the generic relationships of their 
species. 

The problem of separating the restricted genus Triakis from Mustelus 
Linck is discussed, but no solution is seen at present and the two genera are 
considered as one unit for comparison with Hemitriakis. 

A terminology for head morphology, nictitating lower eyelid structure, 
dentition, vertebral groups, and fin morphology is proposed for use with 
carcharhinids. 

LITER.^TURE CITED 
Applegate, Shelton p. 

1965. Tooth terminology and variation in sharks with special reference to the sand 
shark, Carcharias taunts Rafinesque. Los Angeles County Museum of Natural 
History, Contributions in Science, no. 86, pp. 1-18, fig. 1-5. 



94 C ALIFORM A ACADEMY OF SCIENCES [Proc. 4th Ser. 

1967. A survey of shark hard parts. In Sharks, Skates, and Rays, edited by Perry 
W. Gilbert, Robert F. Mathewson, and David P. Rail. Johns Hopkins Press, 
Baltimore, Maryland. Part I, no. 2, pp. .57-67, figs. 1-7, pis. 1-4. 
BiGELOW, Henry B., and William C. Schroeder 

1940. Sharks of the genus Mnstelus in the Western Atlantic. Proceedings of the 

Boston Society of Natural History, vol. 41, no. 8, pp. 417-438, pis. 14-19. 
1944. Nevi? sharks from the Western North Atlantic. Proceedings of the New 

England Zoological Club, vol. 23, pp. 21-36, fig. 1, pis. 7-10. 
1948. Sharks. Fishes of the Western North Atlantic. Memoir 1, Sears Foundation 

for Marine Research, Part 1, Vol. 1, pp. 59-576, figs. 6-106. 

BOULENGER, G. A. 

1902. Description of a new South-African galeid selachian. Annals and Magazine 
of Natural History, ser. 7, vol. 10, pp. 51-52, pi. 4. 
Chen, Johnson T. F. 

1963. A review of the sharks of Taiwan. Department of Biology, College of Science, 

Tunghai University. Biological Bulletin 19, Ichthyological Series no. 1, 
pp. 1-102, figs. 1-28. 
D'Aubrey, Jeannette D. 

1964. Prehminary guide to the sharks found off the east coast of South Africa. South 

African Association for Marine Biological Research, Oceanographic Research 
Institute. Investigational Report no. 8, pp. 1-95, figs. 1-22, pis. 1-28. 

De Buen, Fernando 

1950. El tiburon vitaminico de la costa Uruguaya, Galeorhinus vitaminicus nov. sp., 
y algunas consideraciones generales sobre su Biologia. Servicio Oceanografico 
y de pesca, Montevideo, Urugua}'. Publicaciones Cientificas, no. 4, Contri- 
buciones a la Ictiologia, pp. 152-162, figs. 1-2. 

Fowler, Henry W. 

1929. A list of the sharks and rays of the Pacific Ocean. Proceedings of the Fourth 
Pacific Science Congress, Java, 1929, pp. 481-508. 

1941. Contributions to the biology of the Philippine Archipelago and adjacent 

regions. The fishes of the groups Elasmobranchii, Holocephali, Isospondyli, 

and Ostariophysi obtained by the United States Bureau of Fisheries Steamer 

"Albatross" in 1907 to 1910, chiefly in the Philippine Islands and adjacent 

seas. Bulletin of the United States National Museum, no. 100, vol. 13, pp. i-x, 

1-879, figs. 1-30. 
Garman, Samuel 

1913. The Plagiostoma. Memoirs of the Museum of Comparative Zoology at Harvard 

College, vol. 36, pp. i-xiii, 1-515, pis. 1-77. 
Garrick, J. A. F. 

1954. Studies on New Zealand Elasmobranchii. Part III. A new species of Triakis 

(Selachii) from New Zealand. Transactions of the Royal Society of New 

Zealand, vol. 82, part 3, pp. 695-702, figs. 1-2. 
1967. A broad view of Carcharhinus species, their systematics and distribution. In 

Sharks, Skates and Rays, edited by Perry W. Gilbert, Robert F. Mathewson, 

and David P. Rail. Johns Hopkins Press, Baltimore, Maryland. Part I, no. 5, 

pp. 85-91. 
Garrick, J. A. F., and Leonard P. Sciiultz 

1963. A guide to the kinds of potentially dangerous sharks. In Sharks and Survival, 

edited by Perry W. Gilbert. D. C. Heath and Company, Boston. Pp. 3-60, 

figs. 1-33. 



Vol. XXXVIII] COMPAGXO: SYSTEMATICS OF GENUS HEMITRIAKIS 95 

Gegenbaur, Carl 

1872. Untersuchungen zur Vergleichenden Anatomic der Wirbelthiere. Drittes Heft. 
Das Kopfskelet der Selachier, ein Beitrag zur Erkenntnis der Genese des 
Kopfskeletes der Wirbelthiere. Wilhelm Engelmann, Leipzig. Pp. i-x, 1-316, 
pis. 1-22. 

Gilbert, Perry VV. 

1963. The visual apparatus of sharks. In Sharks and Survival, edited by Perry W. 

Gilbert. D. C. Heath and Company, Boston. Pp. 283-326, figs. 1-30. 
Gilbert, Perry W., and Mark E. Oren 

1964. The selachian nictitans and subocular fold. Copeia, 1964, no. 3, pp. 534-535, 

fig. 1. 
GoHAR, H. A. F., AND F. M. Mazhar 

1964. The Elasmobranchs of the Xorth-Western Red Sea. Publications of the Marine 

Biological Station, Al-Ghardaqa (Red Sea), no. 13, pp. 1-144, figs. 1-81, 

pis. 1-16. 
Herre, Albert W. C. T. 

1923. Notes on Philippine sharks. The Phihppine Journal of Science, vol. 23, no. 1, 

pp. 68-73, pi. 1. 
1953. Check list of Philippine fishes. United States Fish and Wildlife Service, 

Research Report no. 20, pp. 1-977. 

HiLGENDORF, F. 

1904. Ein neuer 5fy///'Hw-artiger Haifisch, Proscyllium habereri nov. subgen., n. spec, 
von Formosa. Sonder-Abdruk aus den Sitzungs-Berichten der Gesellschaft 
naturforschender, Jahrgang 1904, no. 2, pp. 39-41. 
Holmgren, Nils 

1941. Studies on the head in fishes. Embryological, morphological, and phylogenetical 
researches. Part II: Comparative anatomy of the adult selachian skull, with 
remarks on the dorsal fins in sharks. Acta Zoologica, vol. 22, part 1, pp. 
1-100, figs. 1-74. 
Kato, Susumu 

1968. Triakis acutipinna (Galeoidea, Triakidae), a new species of shark from Eucador. 
Copeia, 1968, no. 2, pp. 319-325, figs. 1-2. 
Kato, Susumu, Stewart Springer, and Mary H. Wagner 

1967. Field guide to Eastern Pacific and Hawaiian sharks. United States Fish and 
Wildlife Service, Bureau of Commercial Fisheries, Circular 271, pp. 1-47, 
figs. 1-75. 
Lindberg, G. U., and M. I. Legeza 

1959. Fishes of the Sea of Japan and the adjacent areas of the Sea of Okhotsk and 
the Yellow Sea. Part 1. Amphioxi, Petromyzones, Myxini, Elasmobranchii, 
Holocephali. Academy of Sciences of the Union of Soviet Socialist Republics. 
Keys to the Fauna of the USSR, no. 68, pp. 1-207, figs. 1-108. Translation 
by Israel Program for Scientific Translations, Jerusalem, 1967. 
McCoy, Frederick 

1885. Prodromus of the zoology of Victoria. Melbourne, Australia. Decades 1-10, 
pis. 1-100, pages not numbered. 
MiSRA, K. S. 

1950. On a new species of scyliorhinid fish from Andaman Sea, Bay of Bengal. 
Journal of the Zoological Society of India, vol. 2, pp. 87-90, pi. 1. 
MiYOSi, Y. 

1939. Description of three new species of elasmobranchiate fishes collected at H\'uga 



96 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

Nada, Japan. Bulletin of the Biogeographical Society of Tokyo, vol. 9, pp. 
91-97, figs. 1-3. 
MiJLLER, Johannes, and F. G. J. Henle 

1841. Systematische Beschreibung der Plagiostomen. Berlin. Pp. i-xxii, 1-200, pis. 1-60. 
Norman, J. R. 

1939. Fishes. The John Murray Expedition, 1933-34, Scientific Reports. British 
Museum (Natural History). Vol. 7, no. 1, pp. 1-116, figs. 1^1. 
RiDEWOOD, W. G. 

1921. On the calcification of the vertebral centra in sharks and rays. Philosophical 
Transactions of the Royal Society of London, Ser. B., Zoology, vol. 210, 
pp. 311-407, figs. 1-38. 
Schmidt, P. J. 

1928. On a rare Japanese shark, Calliscyllium venustum Tanaka. Comptes Rendus 

de I'Academie des Sciences de I'URSS, 1928, pp. 65-67, figs. 1-3. 

1929. On two rare Japanese sharks ProscylUum habereri Hilgendorf and Apristuriis 

macrorhynchus Tanaka. Comptes Rendus de I'Academie des Sciences de 
rURSS, 1930, pp. 627-631, figs. 1-2. 
Smith, Hugh M. 

1913. Description of a new carcharioid shark from the Sulu Archipelago. Proceedings 
of the United States National Museum, vol. 45, pp. 599-600, figs. 1-3, pi. 47. 
Smith, J. L. B. 

1957a. A new shark from South Africa. South .\frican Journal of Science, vol. 53, 

no. 10, pp. 261-264, figs. 1-2. 
1957b. A new shark from Zanzibar, with notes on Galeorhinns Blainville. Annals and 
Magazine of Natural History, ser. 12, vol. 10, pp. 585-592, figs. 1-2, pis. 18-19. 
1957c. A preliminary survey of the scylliogaleid dogfishes of South Africa. South 
African Journal of Science, vol. 53, no. 14, pp. 353-359, figs. 1-2. 
Springer, Stewart 

1968. Triakis fehlmanni, a new shark from the coast of Somaha. Proceedings of the 
Biological Society of Washington, vol. 81, pp. 613-624, figs. 1-5. 
Springer, Victor G. 

1964. A revision of the carcharhinid shark genera ScoUodon, Loxodon, and Rhizoprion- 
odon. Proceedings of the United States National Museum, vol. 115, no. 3493, 
pp. 559-632, figs. 1-14, pis. 1-2. 
Springer, Victor G., and J. A. F. Garrick 

1964. A survey of vertebral numbers in sharks. Proceedings of the United States 
National Museum, vol. 116, no. 3496, pp. 73-96, pi. 1. 
Strasburg, Donald W. 

1963. The diet and dentition of Isistius braziliensis, with remarks on tooth replacement 
in other sharks. Copeia, 1963, no. 1, pp. 33-40, figs. 1-5. 
Tanaka, Shigeho 

1912. Figures and descriptions of the Fishes of Japan, including Riukiu Islands, 
Bonin Islands, Formosa, Kurile Islands, Korea, and Southern Sakhalin. 
Tokyo. Vol. 10, pp. 165-186, pis. 41-50. 
Tang, D. S. 

1934. The elasmobranchiate fishes of Amoy. The Natural Science Bulletin of the 
University of Amoy, vol. 1, no. 1, pp. 29-111, figs. 1-22. 
White, E. Grace 

1937. Interrelationships of the elasmobranchs with a key to the order Galea. Bulletin 



Vol. XXXVIIII COMPAGNO: SYSTEMATICS OF GENUS HEMITRIAKIS 97 

of the American Museum of Natural History, vol. 74, no. 2, pp. 25-138, figs. 

1-66, pis. 1-51. 
Whitley, Gilbert P. 

1943a. Ichthyological notes and illustrations (Part 2). The Australian Zoologist, vol. 

10, part 2, pp. 167-187, figs. 1-10. 
1943b. A new Australian shark. Records of the South Australian Museum, vol. 7, 

no. 4, pp. 397-399. 
1944. New sharks and fishes from Western Australia. The Australian Zoologist, vol. 

10, part 3, pp. 252-273, figs. 1-6. 
1948. New sharks and fishes from Western Austraha. Part 4. The Australian Zoologist, 

vol. 11, part 3, pp. 259-2 76, figs. 1-7, pis. 24-25. 



PROCEEDINGS 

OF THE 

CALIFORNIA ACADEMY OF SCIENCES 

FOURTH SERIES 



Festschrift for George Sprague Myers 



Vol. XXXVIII, No. 5. pp. 99-103; 1 fig. December 31, 1970 



NOTES ON THE NATURAL HISTORY 
OF SNIPE EELS 

By 

Giles W. Mead and Sylvia A. Earle 
Harvard University 

One cannot fail to be impressed by the adaptations for midwater life de- 
veloped by the mesopelagic eels, an assemblage doubtless derived from a benthic 
ancestor. In most the body has become far more attenuated than that of the 
most elongate of their benthic relatives, the terminal part of many being fila- 
mentous and apparently composed of little more than minute and poorly ossified 
vertebrae covered by thin skin and supporting fine and hair-like fin rays. This 
attenuation is accomplished by an increase in number of vertebrae rather than 
an increase in the length of each. An apparently intact specimen of N emichthys 
taken during the International Indian Ocean Expedition of 1964 had 670 verte- 
brae — certainly a record number among the vertebrates. Equally extreme are 
specializations in mouth parts. The teeth, for example, vary vastly in form, 
number and position. 

The more extreme genera such as N emichthys, Labkhthys, and Avocettina 
are of concern here. All have greatly prolonged jaws (fig. lb) that bear numer- 
ous small teeth laterally as well as dorsally or ventrally (fig. Id-f), and are 
usually tipped by flattened bony pads that bear teeth or rugosities on all sides. 
The jaw teeth are arranged in chevron-shaped consecutive series or, in others, 
in quincunx. The two halves of the lower jaw are loosely conjoined laterally for 
most of the length of the mandible. The principal part of the upper jaw, in 

[99] 



100 



C ALIFORM A ACADEAIY OF SCIEXCES 



IProc. 4th Ser. 



1 cm 










1 m m 




Vol. XXXVIII] MEAD AND EARLE: SMPE EELS 101 

contrast to other fishes, is composed of the vomer; and the biting elements of 
other lower fishes, the maxillae, are reduced to lateral struts that support the 
base of the prolonged vomerine bar (Beebe and Crane, 1937). The positional 
relationship of these jaws has been particularly enigmatic, for they diverge for- 
ward from the gape so that their tips, and often as much as half of the total 
length of the jaws, cannot be brought into contact with each other when the 
mouth is as far closed as it can be. These fishes can but partially close their 
mouths, yet the distal ends of their jaws, that cannot possibly be brought into 
contact with each other, bear thousands of small chisel-shaped posteriorly in- 
clined teeth (fig. Id-e) reminiscent of the shagreen of an elasmobranch. What 
can be the function of such a structure? 

That these jaws are used to funnel microplankton toward the mouth as the 
eel swims through the water seems unlikely. Lateral movement of the prey by 
but a millimeter or two would take it beyond the grasp of the predator. To feed 
in this way, structural adaptation similar to that in the herrings would be more 
in order. It has also been suggested that these prolonged jaws simply provide 
greater surface area, and might be considered adaptations for flotation. While 
we would consider the attenuated but fin-bearing shape of the body the result 
of selection toward greater surface area that serves the interests of flotation, we 
are reluctant to so consider the development of well ossified structures richly 
endowed with small but dense teeth. Such a beak would also seem to be the 
antithesis of a structure developed in aid of streamlining or locomotion. Nichols 
and Murphy (1944) repeated a report by Mowbray (1922) of a snapper cap- 
tured in Bermuda with a 265 mm. representative of Ncmichthys scolopaceus 
attached by its slender jaws to the posterior margin of the snapper's caudal fin. 
Mowbray concluded, "The specimen being taken in this way gives good reason 
to believe that grasping the tails of fishes is the function of the divergent man- 
dibles of these eels." 

We can suggest an alternative function for these diverging and nonocclusable 
jaws, a suggestion emanating from observations made at mid-depths from the late 
D.S.R.V. Alvin and catches made concurrently by a supporting vessel, R.V. 
Gosnold, both of the Woods Hole Oceanographic Institution. These dives were 
made between October 2 and 6, 1967, in Slope Water of the western North 
Atlantic in an attempt to observe visually certain sound-scattering targets at 



Figure 1. a. Snipe eel, Nemichthys scolopaceus, in typical vertical position as observed 
from the D.S.R.V. Alvin. b. Vertically oriented specimen of A', scolopaceus and serpestid 
shrimp, Sergestes {Sergestes) arcticus, drawn from specimens taken by R.V. Gosnold 
concurrent with D.S.R.V. Alvin observations, c. Distal portion of antenna of S. (S.) 
arcticus. d. Tooth from upper jaw of .V. scolopaceus. e. Inner surface of upper jaw of 
N. scolopaceus. f. Inner surface of lower jaw of A', scolopaceus. 



102 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

mid-depth (see Backus et al., 1968). Concurrent with these dives, R.V. Gosnold 
fished similar depths with a 10-ft. Isaacs Kidd Midwater Trawl. 

Observers aboard Alvin frequently saw snipe eels {N emichthys) at depths 
below 300 m. and confirmed the observations of others that these eels are usually 
oriented vertically in the water, motionless or but slightly undulating, and usually 
with their divergent jaws directed upward. Among the other more spectacular 
animals seen at comparable depths were relatively large sergestid decapods. 
These too were often suspended vertically in the water, their bright orange-red 
stomachs and organs of Pesta prominent, the short pleopods beating furiously, 
and their long antennae extending upward and outward away from the body and 
then turning abruptly to follow a course parallel to the axis of the body to a 
point considerably below the tail. Neither eels nor sergestids appeared to be 
disturbed by the lights of the submarine. 

Both sergestids and snipe eels were caught by the nets of the Gosnold. The 
eels belong to N emichthys scolopaceus Richardson, 1848, and the sergestids, 
kindly identified for us by Mr. Peter Foxton of the National Institute of Ocean- 
ography, Godalming, belong to Sergestes (Sergestes) arcticus Kroyer, 1855. 
Representatives of A' emichthys, as usual, were present in the catch with beaks 
entangled in everything present, living or not. Several were hanging by their 
beaks from the upper part of the netting as the trawl was raised above the water, 
the red stomachs of ingested sergestids visible through the semitransparent 
stomachs and body walls. This material was returned to Woods Hole and to the 
Museum of Comparative Zoology, Harvard University, for study and is deposited 
in the latter institution. 

The stomach contents of about 160 specimens of N emichthys were examined. 
In addition to the Gosnold collection these included others variously collected in 
the western North Atlantic and those from the Indian Ocean and off central 
Chile that were caught during Cruises VI and XIII, respectively, of R.V. 
Anton Bruun. Most stomachs were empty. Those which were not, contained 
crustacean remains exclusively. An examination of the specimens of Sergestes 
which were available and published accounts of others (Burkenroad, 1934, 1937; 
Foxton, 1969; Hardy, 1956) revealed the complexity of the prolonged antennae 
with their multiple sensory hairs, structures admirably suited to aid in flotation. 

We believe that the function of the beak of the snipe eels can be added to 
the list of features in which these eels are unique among vertebrate animals, for 
we suggest here that these animals feed by entanglement. Given the vast extent 
and thread-like nature of some appendages of many mesopelagic crustaceans and 
the set and structure of snipe eel dentition, the evolution of structures adapted 
for the feeding of one on the other seems reasonable. The antennae of a sergestid 
if brushed across the bed of teeth of a snipe eel would almost certainly become 
entangled, and struggle by the prey would only worsen its plight, shorten the 



Vol. XXXVIII] MEAD AND EARLE: SNIPE EELS 103 

distance between shrimp and fish; and ultimately bring the prey within that 
more posterior part of the jaws capable of crushing and swallowing movements. 

Such a feeding mechanism is consistent with present concepts of midwater 
ecology. Food in the deep ocean is scarce and energy precious. Hovering and 
darting, or luring types of predation tend to replace the roving activities more 
prevalent near the surface. Intake per unit of energy expended must be high 
if a predator is to survive. What finer an example of adaptation to these con- 
ditions can there be than that of these eels: hanging effortlessly with flotation 
facilitated through attenuation of the body, and with jaws covered by myriads 
of denticles exquisitely designed to entangle the appendages of passing Crustacea, 
be they moving laterally or, with some, rising or descending as a part of their 
daily routine. 

ACKNOWLEDGMENTS 

This note owes its existence to an invitation to one of us to dive aboard 
Alvin and we thus record our grateful appreciation to the Woods Hole Ocean- 
ographic Institution and especially to Dr. Richard H. Backus of that institution 
for that opportunity. In addition to identifying the sergestids, Mr. Peter Foxton, 
National Institute of Oceanography, Godalming, reviewed the manuscript, as did 
Drs. Backus, R. L. Haedrich, and J. E. Craddock of the Woods Hole Ocean- 
ographic Institution. This paper is contribution number 2351 from the Woods 
Hole Oceanographic Institution. 

REFERENCES 

Backus, R. H., J. E. Craddock, R. L. Haedrich, D. L. Shores, J. M. Teal, A. S. Wlng, 
G. W. Mead, and W. D. Clark 

1968. Ceratoscopelus maderensis: peculiar sound-scattering layer identified with this 

myctophid fish. Science, no. 160, pp. 991-993. 
Beebe, W., and J. Crane 

1937. Deep-sea fishes of the Bermuda Oceanographic Expeditions. Family Nemichthy- 
idae. Zoologica (N.Y.), vol. 22, no. 4, pp. 349-383. 
Burkenroad, M. D. 

1934. The Penaeidae of Louisiana with a discussion of their world relationships. Bulletin 

of the American Museum of Natural History, vol. 68, no. 2, pp. 61-143. 
1937. The Templeton Crocker E.xpedition XII. Sergestidae (Crustacea, Decapoda) from 
the Lower California region, with descriptions of two new species and some 
remarks on the organs of Pesta. Zoologica (N.Y.), vol. 22, no. 4, pp. 315-529. 
Foxton, P. 

1969. The morphology of the antennal flagellum of certain of the Penaeidae (Decapoda, 

Natantia). Crustaceana, vol. 16, no. 2, pp. 33-42, 1 pi. 
Hardy, A. C. 

1956. The open sea. Boston, xv + 335 pages, 18 pis. 
Mowbray, L. L. 

1922. Habit note on snipe eel. Copeia no. 108, page 49. 
Nichols, J. T., and R. C. Murphy 

1944. A collection of fishes from the Panama Bight, Pacific Ocean. Bulletin of the 
American Museum of Natural History, vol. 83, no. 4, pp. 217-260. 



PROCEEDINGS 

OF THE 

CALIFORNIA ACADEMY OF SCIENCES 

FOURTH SERIES 



Festschrift for George Sprague Myers 



Vol. XXXVIII, No. 6, pp. 105-130; 7 figs.; 11 tables December 31, 1970 



THE ZOOGEOGRAPHY 

OF THE HERPETOFAUNA 

OF THE PHILIPPINE ISLANDS, 

A FRINGING ARCHIPELAGO 

By 

Walter C. Brown 

and 

Angel C. Alcala 

Division of Systematic Biology, Stanford Uniziersity and Menlo College, 

JMenlo Park, California and 

Department of Biology, Silliman University, Philippines 

INTRODUCTION 

Inger, in his essay on the zoogeography of the PhiHppine amphibia (1954, 
pp. 448-510), presented the first major distributional paper for any part of the 
herpetofauna since Taylor's essay (1928). The first part of Inger's paper is 
concerned with the geological history of the Philippines, and the origins and 
degree of endemism exhibited by the amphibian fauna. Secondly, he discusses 
the pathways of entry into the Philippines in terms of the location of the nearest 
relatives and possible time of entry into the Philippines. He notes, for example, 
that the present distribution of the genus Platymantis (replaces Cornufer, 
Zweifel, 1967) suggests two speciation centers, one in the New Guinea-Solomons 
region and one in the northern Philippines; but by analogy, in comparison with 
some other amphibians, suggests a Papuan origin and subsequent dispersal into 
the Philippines. Inger therefore regards these two present centers as peripheral 
isolated concentrations of a once more widely distributed genus (1954, pp. 494, 

[105] 



106 CALIFORX I A ACADEMY OF SCIENCES [Proc. 4th Ser. 

497). He suggests that the bulk of the amphibians entered the PhiHppines by 
2 major routes, the Palawan or Sulu-Mindanao routes. He discusses relative 
time of entry of different components of the amphibian fauna primarily in terms 
of extent of endemism and distance from areas occupied by presumed nearest 
relatives. Within the Philippines, Inger recognizes only 2 somewhat doubtful 
zoogeographic subdivisions. In his discussion of dispersal (pp. 475-484), he 
notes that both dispersal by way of earlier land connections and over-water 
dispersal must be considered, and he also notes, in general descriptive terms, 
possible routes within the Philippines. 

Leviton (1963) provides the most recent discussion of zoogeography of the 
terrestrial snake fauna of the Philippine archipelago. His discussion is primarily 
concerned with extraterritorial origins, time of entry and endemism, present 
distributions, and the taxonomic relationships of species within the Philippines. 
These are considered in terms of past changes in island configurations, and 
probable internal pathways. He states (p. 377), contrary to Inger's views 
relative to the dispersal of the amphibians (Inger, 1954, p. 484), that the present 
distribution of the snakes can, for the most part, be explained on the basis of 
former land connections. He recognizes 5 faunal (serpentine) subregions within 
the archipelago at the present time. 

Both authors very ably discuss the present distribution of the faunal element 
with which they are concerned in terms of traditional concepts of extraterritorial 
origins, pathways of entry and internal dispersal as governed by probable 
geological changes, time of entry, and means of dispersal. 

Darlington (1957, pp. 476-541) discusses the Philippine ichthyological and 
herpetological faunas in the more general context of distributions on fringing 
archipelagoes. Immigrant patterns of distribution, where the species are distrib- 
uted along the migration route with dropouts occurring linearly as determined 
by distance and relative dispersal abilities, are, he believes, the primary patterns 
exhibited in fringing archipelagoes. This basic pattern is modified for older 
relict groups by concentration of species on distal or proximal islands within 
the archipelago (p. 533). 

MacArthur and Wilson (1963) propose the hypothesis that the number of 
species on an island represents a balance between number of species reaching 
the island and number of species becoming extinct per unit of time. They point 
out that a number of interacting variables will determine the point at which 
these 2 curves intersect. These include distance from source of immigrants, the 
species pool of immigrants, area of island concerned, or some other limiting factor 
such as unfavorable climate. They further propose that in time secondary radia- 
tion centers should increase with distance of islands from the major source of the 
fauna, when corrections are made for area or other limiting variables. They also 
note that the number of species decreases more rapidly for large than for small 
islands with increasing distance from source of colonization. 



Vol. XXXVIII] BROWX AXD ALCALA : HERPETOFAUNA OF PHILIPPINES 107 





3 < 
•o .S 



-v -5 



M -^ 






=a 



PROBLEMS AND METHODS 

Recent intensive exploration of the herpetofaunas of 5 Philippine islands, 
Palawan, Mindanao, Bohol, Negros, and Mindoro (fig. la), makes possible more 
critical examination of many of the zoogeographic hypotheses suggested by 
previous authors. Factors which we believe make this possible are: 1) intensive 
sampling techniques which provide more accurate estimates of species-diversity; 
2 ) choice of islands from proximal, distal, and intermediate regions of the 
archipelago (fig. la); 3) range in size from 5,000 to 95,000 square kilometers 



108 CALIFORNIA ACADEMY OF SCIENCES LProc. 4th Ser. 

Table 1. Intensively explored areas on the five islands included in the recent survey. 



Mountain Region 


A Ititudc 
(meters) 


Island 




Island Area 
(sq. km.) 


Exploration Date 


Cuernos de Negros 


1,903 


Negros 
(southern part) 




12,700 


March — May 

1958 

(about 7 weeks) 


Mt. Canlaon 


2,463 


Negros 
(northern part) 




" 


March — April 

1962 

(about 4 weeks) 


Dapitan Peak 


2,199 


Mindanao 
(Zamboanga Peninsula) 


94,600 


March — May 

1959 

(about 6 weeks) 


Thumb Peak 


1,286 


Palawan 
(central part) 




11,800 


April — May 

1961 

(about 7% weeks) 


Mt. Halcon 


2,580 


Mindoro 
(northern part) 




9,750 


April — May 

1963 

(about 4 weeks) 


Sagungan Mountain 


870 


Bohol 
(southeastern pa 


rt) 


4,100 


April — May 

1962 

(about 4 weeks) 



(table 1); 4) choice of islands which encompass sufficiently large areas of 
original and/or secondary lowland forest as to make negligible differences in 
diversity which might be due to major differences in the dominant type of plant 
community (see Brown and Alcala, 1964). 

The techniques stressed intensive sampling of arboreal, surface, and sub- 
terranean strata in the lowland forest whenever possible, as well as selected 
mountains. The expeditions to each of the 6 mountain areas on the 5 islands 
were carried out by crews of 8 to 10 men over 4 to 7V:i week periods (table 1). 



Table 2. Number of species recorded for the islands included in this study. The number 
in parentheses is the number of species belonging to the group of 23 widely distributed species 
associated with man's economy or beach communities. The number in brackets is the number 
of relict species. 



Palawan 



Mindanao 



Bohol 



Negros 



Mindoro 



Leyte 



Luzon 



34(4)[3] 21(3 )[4] 16 ( 4 )[ 4 ] 12(4)[2] 16 (3)[3] 2l(4)[6] 
52(10)[6] 31(11)[4] 33 (11)[ 4] 27 (ll)[2] 34 ( 8 )[ 7 ] 

39(9)[2] 20(8)[1] 29(9)[2] 19(8)[l] 18(7)[l] 39 ( 9 )[ 3 ] 



Caecilians 




1 


Frogs 


22 1 


(4)[1] 


Lizards 


23 


(11)[1] 


Snakes 


33 


(9)[0] 



Total 78(24)[2] 125 (23)[ ll] 72 (22)[9] 7S(24)[lO] 58 (23)[5] 



94(21)[15] 



Vol. XXXVIII] BROWN AND ALCALA : HERPETOFAUNA OF PHILIPPINES 109 

Table 3. Ainphibiaus kiioivn from Palawan, Mindanao, Bohol, Negros, Mindoro, Leyte, 
and Luzon islands. 







Palawan 


Mindanao 


Bohol 


Negros 


Mindoro 


Leyte 


Luzon 


1. 


Ichthyophis monochrous 


X 


X 












2. 


Barbourula busuangensis 


X 














3. 


Ansonia mcgregori 




X 












4. 


Ansonia mulleri 




X 












5. 


Bufo biporcatiis 


X 














6. 


Leptobrachium hasselti 


X 


X 


X 




X 






7. 


Megrophrys monticola 


X 


X 


X 






X 




8. 


Platymantis cormdiis 














X 


9. 


Platymantis corrugatns 




X 


X 


X 


X 


X 


X 


10. 


Platymantis dorsalis 




X 


X 


X 


X 


X 


X 


11. 


Platymantis guentheri 




X 


X 


X 




X 


X 


12. 


Platymantis hazelae 








X 






X 


13. 


Platymantis ingeri 






X 










14. 


Platymantis subterreslris 














X 


IS. 


Micrixalus mariae 


X 














16. 


Ooeidozyga diminutiva 




X 












17. 


Ooeidozyga laevis 


X 


X 


X 


X 


X 


X 


X 


18. 


Rana cancrivora 


X 


X 


X 


X 


X 


X 


X 


19. 


Rana erythraea 








X 








20. 


Rana everetti 




X 




X 




X 


X 


21. 


Rana leytensis 




X 


X 


X 




X 




22. 


Rana limnocharis 


X 


X 




X 


X 




X 


23. 


Rana magna 


X 


X 


X 


X 


X 


X 


X 


24. 


Rana microdisca 


X 


X 












25. 


Rana nicobariensis 


X 














26. 


Rana sanguinca 


X 












.,1 


27. 


Rana signata 


X 


X 


X 




X 


X 


X 


28. 


Rana woodworthi 














X 


29. 


Staurois natator 


X 


X 


X 






X 




30. 


Philautus aciitirostris 




X 












31. 


Philautus bimaculatus 




X 


X 










32. 


Philautus leitensis 






X 






X 




3i. 


Philautus longicrus 


X 














34. 


Philautus pictus 


X 














35. 


Philautus schmackeri 










X 






36. 


Philautus spinosus 




X 


X 










37. 


Philautus williamsi 














X 


38. 


Rhacophorus appendiculat us 




X 


X 


X 








39. 


Rhacophorus everetti 


X 














40. 


Rhacophorus emembranatus 




X 












41. 


Rhacophorus leucomystax 


X 


X 


X 


X 


X 


X 


X 


42. 


Rhacophorus lissobrachius 




X 


X 










43. 


Rhacophorus pardalis 




X 


X 


X 






X 


44. 


Rhacophorus surdus 




X 










X 


45. 


Pelophryne albotaeniata 


X 















110 CALIFORMA ACADEMY OF SCIENCES [Proc. 4th Ser. 

Table 3. Continued. 







Palmvan 


Mindanao 


Bo hoi 


Negros 


Mindoro 


Leyte 


Luzon 


46. 


Pelophryne brevipes 




X 












47. 


Pelophryne lighti 




X 


X 










48. 


Chaperina fiisca 


X 


X 












49. 


Kalophrynus plenrostigma 




X 


X 






X 




50. 


Kaloula baleata 


X 












X 


51. 


Kaloula conjuncta 




X 




X 


X 


X 


X 


52. 


Kaloula picta 




X 




X 


X 


X 


X 


5i. 


Kaloula rigid a 














X 


54. 


Oreophryne annulata 




X 













The belief that our samphng techniques for these selected mountain areas has 
provided a realistic estimate of the number of species present in the area is based 
upon experience in the Cuernos de Negros area in southern Negros Island. The 
initial survey expedition there in 1958, using the sampling methods noted above, 
resulted in the recording of 67 herpetofaunal species (Brown and Alcala, 1961, 
p. 631). Although extensive resampling of this area has occurred during the 
interim of several years, in connection with our population and other ecological 
studies, only 7 additional species {Hemiphyllodactylus typus, Lepidodactylus 
christiani, L. lugubris, Luperosaurus cumingi, Brachymclcs tridactylus, Typhlops 
cumingi, and Boiga angulata) have been found in the southern Negros area. 
The 4 remaining species in the present list for Negros (tables 3, 4 and 5) are 
known only from the northern part of the island. Thus, even though Bohol, 
Palawan, and Mindoro are not yet widely explored, the presently available data 
on their herpetofaunal communities, based on past records and on our intensive 
sampling of selected mountain areas, are believed to be sufficient to realistically 
approximate their relative positions in terms of diversity. 

Utilizing primarily the data from the 6 recently intensively explored mountain 
areas on 5 islands (table 1), supplemented by available lists of species, largely 
based on data assembled by earlier explorers, for Leyte and Luzon islands as 
well as total distributional data for a few genera, we propose to examine: 1) the 
nature of the relict patterns for some of the older herpetofaunal elements; 2) the 
probability of secondary radiation zones; 3) the relative importance of probable 
internal migration routes, using Sorenson's index of similiarity; 4) the probable 
importance of marine barriers in effecting the present distribution patterns; 
5 ) the relation of island size and distance to diversity of species. 

RESULTS 

Diversity of Species 
A total of 9,000+ herpetofaunal specimens, ranging from about 1,000 for 
Mindoro Island to 2,500 for Negros Island, were collected during our recent 
expeditions. Classification of the collected material when added to earlier rec- 



Vol. XXXVIin BROWN AND ALC ALA : HERPETOFAUXA OF PHILIPPIXES 111 

ords, reveals 197 species: 1 caecilian, 49 frogs, 82 lizards, and 66 snakes for these 
five islands (tables 2-5). Although the new records ranged from 61 for Bohol 
to 5 for Mindoro, only 14 were new species or species not recorded from the 
Philippines prior to our intensive sampling. Most of these 14 were from Palawan 
and ^Mindanao, the islands adjacent to Borneo. 

The number of species of snakes and frogs recognized from Luzon and Lej^te 
islands are based primarily upon Inger's review of the Philippine amphibia 
(1954) and Leviton's review of the snakes (1963). The distribution of the 
species of the genus Calamaria, however, is from Inger and Marx (1965). The 
list of lizards for Luzon is based primarily on the earlier records of Taylor ( 1922a, 
1922b, 1922c, 1923 and 1925) with a few recent additions by the present authors. 
The inclusion of Luzon adds 23 more species (5 amphibians, 10 lizards, and 8 
snakes) to the 197, making a total of 220 species (tables 3-5). 



Native Fauna 

Twenty-three of the 220 species (4 frogs, 11 lizards, and 8 snakes) are re- 
garded as probably nonnative; that is, as possibly introduced or at least re- 
introduced by man. We do not presume that this list includes all species which 
have at any time been introduced by man, intentionally or otherwise. It probably, 
however, does include most of these species which are readily, perhaps often 
accidentally, transported from island to island. To be included in this category, 
the species must meet these criteria: 1 ) occur on at least 4 of the 5 islands 
included in our study; 2) exhibit no subspeciation except for the Palawan 
populations in some instances; 3) be associated with man's habitations, and 
cultivated lands, or with other lowland beach communities (Brown and Alcala, 
1964); 4) be widespread in Borneo and other adjacent areas. It is interesting 
to note (table 2) that the number of species of this nonnative group, whether we 
are concerned with frogs, lizards, or snakes, is essentially the same for each of the 
7 islands irrespective of distance from entry-point or area of island. This we 
interpret as further evidence supporting their classification in this category. 
The 23 species include: 



Amphibians: 
Ooeidozyga laevis 
Rana cancrivora 
Rana limnocharis 
Rhacophorus leucomystax 



Lizards: 

Cosymbottis platyiirus 
Gehyra mutilata 
Gekko gecko 
Hemidactylns frenatus 
Hemiphyllodactylus typus 
Draco volans 
Dasia smaragdina 
Emoia atrocostata 
Lygosoma (Leiolopisma) 

quadrivittatum 
Mabuya multifasciata 
Mabiiva midticarinata 



Snakes: 

Typhlops braminae 
Python reticulatus 
Ahaetidla prasina 
Dendrelaphh pic t us 
Chrysopelea paradtsi 
Elaphe erythrura 
Lycodon aulicus 
Psammodynastes pidvendentus 



112 CALIFORMA ACADEMY OF SCIENCES LProc. 4th Ser. 

Table 4. Lizards known from Palaivan, Mindanao, Bohol, Negros, Mindoro, and Luzon 
islands. 



Palawat 


I Mindanao 


X 


X 




X 




X 


X 




X 


X 


X 




X 


X 


X 


X 


X 




X 


X 



Bohol Negros Mindoro Luzon 



1. Cosymbotus platyurus X X X X X X 

2. Cyrtodactyhis agusanensis 

3. Cyrtodactyhis annidatus 

4. Cyrtodactyhis philippinicus X 

5. Cyrtodactylus redimicidns 

6. Gehyra mutilata 

7. Gekko athymus 

8. Gekko gecko 

9. Gekko mindorensis 

10. Gekko monarchus 

11. Gekko palawanensis 

12. Hemidactyhis jrenatus 

13. Hemidactyhis garnoti 

14. Hemidactylus luzonensis 

15. Hemiphyllodactylus typits X XXX 

16. Lepidodactyhis aiireolineatus X X 

17. Lepidodactyhis christiani X 

18. Lepidodactyhis herrei X 

19. Lepidodactyhis higubris X 

20. Lepidodactyhis natijanensis X 

21. Lepidodactyhis planicaiidus X X 

22. Luperosauriis cumingi X 

23. Luperosaiirus joloensis X 

24. Perochints ateles 

25. Pseudogekko compressicorpiis XX X 



X 




X 


X 




X 


X 




X 


X 



X 


X 


X 




X 


X 


X 


X 




X 



26. Pseudogekko brevipes X X 

27. Ptychozoon intermedia 

28. Calotes cristateUus 

29. Calotes marmoratus X X 

30. Draco bimaculatus X 

31. Draco everetti X 

32. Draco mindanensis 

33. Draco ornatiis 

34. Draco quadrisi 

35. Draco rizali 

36. Draco volans 

37. Gonyocephalus interrupt us 

38. Gonyocephalus semperi 

39. Gonyocephalus sophiae 

40. Hydrosaurus pustulosus 

41. Varanus salvator 

42. Dibamus argenteus 

43. Brachymeles bonitae 

44. Brachymeles elerae 

45. Brachymeles gracilis 

46. Brachymeles pathjinderi 





X 






X 


X 




X 


X 








X 


X 


X 


X 


X 
X 


X 




X 


X 










X 




X 








X 


X 


X 


X 


X 


X 


X 




X 


X 


X 


X 


X 




X 


X 


X 
X 




X 


X 


X 


X 


X 




X 











Vol. XXXVIII] BROWN AND ALCALA : HERPETOFAUNA OF PHILIPPINES 113 



Table 4. Continued. 



Palawan Mindanao Bohol Ne.gros Mindoro Luzon 



47. 
48. 
49. 
SO. 
51. 
52. 
Si. 
54. 
55. 
56. 
57. 
58. 
59. 
60. 
61. 
62. 
63. 
64. 
65. 
66. 
67. 
68. 
69. 
70. 
71. 
72. 
73. 
74. 
75. 
76. 
77. 
78. 
79. 
80. 
81. 
82. 
83. 



87. 
88. 
89. 
90. 
91. 
92. 



Brachymeles samarensis 
Brachymeles schadcnhergi 
Brachymeles talinis 
Brachymeles tridactylus 
Brachymeles wrighti 
Brachymeles hilong 
Dasia griffini 
Dasia olivaceum 
Dasia smaragdina 
Emoia atrocostata 
Emoia caeruleocauda 
Emoia ruficauda 

Lygosoma {Leiolopisma) auriculatmu 
Lygosoma {Leiolopisma) pulchellum 
Lygosoma (Leiolopisma) quadrivittatum 
Lygosoma (Leiolopisma) rabori 
Lygosoma (Leiolopisma) semperi 
Lygosoma (Leiolopisma) subvittatum 
Lygosoma (Leiolopisma) vulcanium 
Lygosoma (Leiolopisma) zamboangensis 
Lygosoma (Lygosoma) chalcides 
Lygosoma (Sphenomorphus) acntum 
Lygosoma (Sphenomorphus) arborens 
Lygosoma (Sphenomorphus) atrigularis 
Lygosoma (Sphenomorphus) coxi 
Lygosoma (Sphenomorphus) decipiens 
Lygosoma (Sphenomorphus) diivati 
Lygosoma (Sphenomorphus) jasciatum 
Lygosoma (Sphenomorphus) jagori 
Lygosoma (Sphenomorphus) luzonensis 
Lygosoma (Sphenomorphus) mindanensis 
Lygosoma (Sphenomorphus) palawanensis 
Lygosoma (Sphenomorphus) steerei 
Lygosoma (Sphenomorphus) stejnegeri 
Lygosoma (Sphenomorphus) varigatum 
Lygosoma (Sphenomorphus) wrighti 
Lygosoma (Sphenomorphus) sp. 
Mabuya bontocensis 
Mabuya multicarinata 
Mabuya multifasciata 
Otosaurus cumingi 
Tropidophoriis grayi 
Tropidophorus leucospilos 
Tropidophoriis misaminus 
Tropidophorus partelloi 
Tropidophorus sp. 



X 



X 



X 


X 


X 


X 


X 



X 






X 


X 


X 


X 


X 


X 


X 


X 


X 


X 








X 












X 


X 


X 


X 


X 




X 


X 


X 
X 




X 








X 








X 








X 











X 


X 


X 


X 


X 
X 




X 


X 








X 














X 


X 


X 


X 


X 
X 




X 


X 








X 










X 

X 


X 


X 


X 


X 


X 


X 


X 


X 


X 


X 


X 


X 




X 


X 


X 


X 


X 
X 
X 




X 












X 












X 











114 CALIFORMA ACADEMY OF SCIENCES LProc. 4th Ser. 

Table 5. Snakes known from Palawan, Mindanao, Bohol, Negros, Mindoro, Leyte, and 
Luzon islands. 







Palawan 


Mindanao 


Bohol 


Negios 


Mindoro 


Leyte 


Luzon 


1. 


Typhlops braminae 


X 


X 


X 


X 




X 


X 


2. 


Typhlops ciimingi 








X 








3. 


Typhlops canlaonensis 








X 








4. 


Typhlops dendrophis 




X 












5. 


Typhlops jagori 














X 


6. 


Typhlops longicauda 




X 


X 










7. 


Typhlops luzonensis 








X 






X 


8. 


Typhlops mindanensis 




X 












9. 


Typhlops ruber 










X 






10. 


Typhlops ruficauda 














X 


11. 


Typhlops rugosa 




X 












12. 


Xenopeltis unicolor 


X 














13. 


Python reticidatus 


X 


X 


X 


X 


X 


X 


X 


14. 


Ahaetidla prasina 


X 


X 


X 


X 




X 


X 


IS. 


Aplopeltura boa 


X 


X 












16. 


Calamaria bitorques 














X 


17. 


Calamaria gervaisi 




X 




X 


X 




X 


18. 


Calamaria lumbricoidea 




X 


X 






X 




19. 


Calamaria palawanensis 


X 














20. 


Calamaria virgulata 


X 


X 












21. 


Chrysopelea paradisi 


X 


X 




X 


X 


X 


X 


22. 


Cydocorns lineatus 




X 




X 


X 




X 


2i. 


Dendrelaphis candolineatus 


X 


X 


X 


X 


X 


X 


X 


24. 


Dendrelaphis pictus 


X 


X 


X 


X 


X 




X 


25. 


Dryophiops philippina 








X 


X 




X 


26. 


Dryocalamus tristrigat us 


X 














27. 


Dryocalamiis subannidatus 


X 














28. 


Elaphe erythrura 


X 


X 


X 


X 


X 


X 


X 


29. 


Gonyosoma oxycephala 


X 




X 


X 


X 




X 


30. 


Hologerrhum philippinum 














X 


31. 


Hurria rynchops 


X 


X 




X 




X 


X 


2,2. 


Liopeltis philippinus 


X 














33. 


Liopeltis tricolor 


X 














34. 


Lycodon aidicus 


X 


X 


X 


X 


X 




X 


35. 


Lycodon dumerili 




X 












36. 


Lycodon mulleri 










X 




X 


37. 


Lycodon subcinctus 


X 














38. 


Lycodon tesselatus 














X 


39. 


Myersophis alpetris 














X 


40. 


Natrix auriculata 




X 


X 






X 




41. 


Matrix chrysarga 


X 














42. 


Natrix dendrophiops 




X 


X 


X 


X 




X 


43. 


Natrix lineata 




X 












44. 


Natrix spilogaster 














X 


45. 


OUgodon ancorus 










X 




X 



Vol. XXXVIIIJ BROWX AXD ALCALA : HERPETOFAUNA OF PHILIPPINES 115 



Table 5. Continued. 



Palawan Mindanao Boliol Negros Mindoro Leyte Luzon 



46. 


Oligodon niacidatus 




X 








47. 


OUgodon modestum 




X 




X 




48. 


Oligodon vertebralis 


X 


X 








49. 


Opisthotropis alcalai 




X 








50. 


Opisthotropis typica 


X 










51. 


Oxyrhabdiuin leporinum 








X 


X 


52. 


Oxyrhabdium modestum 




X 


X 






53. 


Psammodynastes pidvendentus 


X 


X 


X 


X 




54. 


Pseiidorabdion ater 




X 








55. 


Pseudorabdion mcnamarae 








X 




56. 


Pseudorabdion montaniim 








X 




57. 


Psetidoyabdion oxycephalum 








X 




58. 


Pseudorabdion taylori 




X 








59. 


Sibynophis bivittatus 


X 










60. 


Stegonotus mi'dleri 




X 








61. 


Zaocys carinatus 


X 










62. 


Zaocys hizonensis 








X 




63. 


Bioga angulata 




X 


X 


X 




64. 


Bioga cynodon 


X 


X 








65. 


Boiga dendrophila 


X 


X 


X 






66. 


Boiga drapiczi 


X 










67. 


Boiga philippina 












68. 


Calliophis calligaster 








X 


X 


69. 


Maticora intestinalis 


X 


X 


X 






70. 


Naja naja 


X 


X 


X 




X 


71. 


Ophiophagiis hanna 


X 


X 




X 


X 


72. 


Trimeresurus jlavomacidat us 




X 


X 


X 


X 


73. 


Trinieresiirus schidtzei 


X 










74. 


Trimeresurus wagleri 


X 


X 


X 


X 





X 



X 


X 


X 


X 


X 


X 




X 




X 




X 




X 


X 


X 




X 


X 


X 



In other sections of this paper, certain of the indices are determined, both 
for the total fauna and for the presumed native fauna following the exclusion 
of these 25 species. 

Relict Patterns and Secondary Radiation Zones 

As noted by Darlington (1957, p. 505) endemism at the specific and sub- 
specific levels is high, but at the generic level very low for the Philippine herpeto- 
fauna. It is of interest to examine the distributional patterns of these endemic 
genera, as well as of a few other genera (the presumed earlier arrivals) which, 
though not limited to the Philippines, exhibit disrupted distributional patterns 
outside the Philippines, to determine their fit to typical relict or modified im- 
migrant patterns as postulated by Darlington (1957, p. 484 ff.). This selection 
of endemic genera and those with strongly disrupted distributional patterns does 



116 



CALIFORNIA ACADEMY OF SCIENCES 



[Proc. 4th Ser. 



* 555955555555 







5 < 5 .^ - I J 
^ a o ? p o 2 

u) ij .*^ -!0 -12 1? -^ 

c c c S S ^ S 

o o o o 5 o g 

6 6 6 I I 6 I 

:v ^ i. X ^ X =^ 




Q Q 






Vol. XXXVIII 1 BROWN AND ALCALA : HERPETOFAUNA OF PHILIPPINES 117 

not preclude the possibility that species in other, widely distributed genera may 
also be relicts of early immigrations, but provides objective criteria for selection. 

The endemic genera include Barbourula, an amphibian; Luperosawus, 
Pseudogekko and Brachymeles, lizards; and Cyclocorus, Hologerrhiim, and 
Oxyrhabdiuin, snakes. Platymantis in the amphibia and Perochirus in the 
squamata, though not endemic, exhibit overall disrupted patterns. 

Barbourula and Perochirus are limited to single species in the Philippines. 
Barbourula is known only from the Palawan group and Perochirus only from 
Mindanao and Leyte. This pattern of limitation to proximal islands permits 3 
alternative explanations : 1 ) these genera have very low dispersal ability and 
did not go beyond these proximal islands; 2) they have reached but failed to 
establish themselves on more distal islands; 3) they still remain to be found 
on the other islands. 

Platymantis, Brachymeles, and Luperosaurus exhibit relict patterns (figs. 2, 
3, and 4) of the type which may be interpreted as resulting from the partial 
extinction of an old widespread fauna (Darlington, 1957. p. 485). Each of 
these genera includes several species. The genus Platymantis (as noted by Inger, 
1954, p. 496 exhibits a concentration of species on Luzon at the distal end of 
the archipelago, others with distribution limited to one or two scattered islands, 
and several species which are widespread throughout the Philippines, though 
none of these species are known from outside of the archipelago. As proposed by 
MacArthur and Wilson (1963, p. 386) and Inger (1957, p. 496), such a pattern 
may also be interpreted as due in part to secondary radiation from the distal 
island of Luzon as indicated by the present distribution of P. hazelae and P. 
poUllensis, very closely related species, not necessarily as wholly due to chance 
extinctions of once widespread species. 

The genus Brachymeles, as reviewed by Brown and Rabor (1967), includes 
such widespread species as B. gracilis and B. schadenbergi, groups of species 
limited to the proximal or distal islands, and a third very interesting group (the 
tridactylus-vermis group) which strongly suggests an origin in and radiation from 
the center of the archipelago. Based on the presumed evolutionary relationships, 
indicated by the degree of reduction of the limbs and digits, the least specialized 
members of the group occur in the central islands and those with the greater 
reduction of limbs and digits in the northern and southern islands. These species 
include: 









Digits 


on 


Digits on 


Species 


Distribution 




fore limbs 


hind limbs 


Brachymeles tridactylus 


Negros 




3 




3 


Brachymeles cebuensis 


Cebu 




3 




2 


Brachymeles samarensis 


Leyte, Samar, 


and Luzon 


2 




2 


Brachymeles bonitae 


Luzon and Mi 


indoro 


0-1 




0-1 


Brachymeles vermis 


Sulus 













118 



CALIFORNIA ACADEMY OF SCIENCES 



[Proc. 4th Ser. 









%. 



■q. o 






O 



Q Q 



w 


w 


ai 


« 


D 


t:> 


o 


O 



Vol. XXXVIII] BROWX AXD ALCALA : HERPETOFAUNA OF PHILIPPINES 119 

Table 6. Indices of similarity for amphibians. The index in parentheses is based on the 
native species after exclusion of common widespread forms associated with man's economy 
or beach communities. The index in brackets is based on those species remaining after the 
relicit species are also excluded. 





Mindanao 


Leyte 


Bo ho I 


Negros 


Mindoro 


Luzon 


Palawan 


0.429 


0.368 


0.372 


0.263 


0.412 


0.326 




(0.333) 


(0.258) 


(0.278) 


(0.067) 


(0.231) 


(0.171) 




[0.364] 


[0.296] 


[0.323] 


[0.080] 


10.261] 


[0.214] 


Mindanao 




0.600 


0.691 


0.520 


0.476 


0.509 






(0.588) 


(0.667) 


(0.429) 


(0.368) 


(0.426) 






[0.486] 


10.634] 


[0.343] 


[0.303] 


[0.368] 


Leyte 






0.703 


0.687 


0.643 


0.595 








(0.645) 


(0.640) 


(0.571) 


(0.533) 








[0.583] 


[0.556] 


[0.500] 


[0.476] 


Bohol 








0.541 
(0.467) 
[0.364] 


0.485 
(0.385) 
[0.300] 


0.429 
(0.343) 
[0.240] 


Negros 










0.643 
(0.500) 
[0.429] 


0.703 
(0.621) 
[0.526] 


Mindoro 












0.606 
(0.480) 
[0.471] 



Hologerrhum has a relict pattern exhibiting hmitation to the distal island of 
Luzon (fig. 5). Pseudogekko, with two species, has a spotty distribution; 
Cyclocorus (1 species) and Oxyrhabdium (2 species) have less obvious relict 
patterns, being more widely distributed throughout the archipelago (figs. 4 and 

5). 

Internal Migration Routes 

When differences in comiDOsition as well as diversity are considered, it is 

possible to compare the relative effectiveness of probable disiDersal routes 

indicated in fig. lb. In comparing composition of the faunas, we have used the 

twice the number of species common to the two communities 

similarity index — -. ; ,: — —. — 

the sum of the species comprismg each of the communities 

developed by Sorenson (1948) for comparing plant communities in northeast 

Greenland and in Denmark. A low index of similarity will be the result of either: 

(1) a large difference in diversity; or (2) when diversities are more or less 

equal, a small number of species in common. 

In addition to the 5 islands upon which this study is primarily based, Lej^te 

Island, as well as Luzon, has been included in this section since the former lies 



120 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

Table 7. Indices of similarity for lizards. (The index in parentheses is based on the 
species exclusive of the 23 widespread forms associated ivith man's economy or beach 
communities, that in brackets is based on the species remaining after relicit species are also 
excluded). 





Mindanao 


Bo ho I 


Negros 


Mindoro 


Luzon 


Palawan 


0.400 


0.481 


0.500 


0.440 


0.316 




(0.185) 


(0.125) 


(0.176) 


(0.000) 


(0.053) 




[0.213] 


[0.148] 


[0.207] 


[0.000] 


[0.067] 


Mindanao 




0.651 


0.471 


0.380 


0.395 






(0.548) 


(0.313) 


(0.207) 


(0.265) 






[0.539] 


[0.333] 


[0.200] 


[0.218] 


Bohol 






0.562 


0.552 


0.491 








(0.333) 


(0.278) 


(0.348) 








[0.294] 


[0.267] 


[0.286] 


Negros 








0.600 
(0.368) 
[0.375] 


0.478 
(0.333) 
[0.323] 


Mindoro 










0.590 
(0.429) 
[0.424] 



closest to Mindanao on the most eastern dis[3ersal route. The Leyte indices have 
been computed only for frogs and snakes, based on lists published by Inger 
(1954) and Leviton (1963). The lizards are not sufficiently well known to be 
included at this time, and the snakes are believed to be rather poorly known 
since several widely distributed species have not yet been reported from this 
island. This, however, would tend to introduce an error in the direction of a 
low rather than a high index. 

As is evidenced in table 2, Mindanao, at the proximal end of the eastern 
routes, exhibits a higher diversity for each of the major taxa considered, as well 
as for the herpetofauna as a whole, than do any of the other 4 islands included 
in our expeditions. This will depress the similarity index when comparisons are 
made with the other islands, whereas no such depressing effect will exist in the 
instance of Palawan at the proximal end of the western entryway, since the 
diversity of the Palawan herpetofauna is about the same as that of the more distal 
islands which range from 60 to 80 species in total herpetofauna. Thus, if indices 
between Mindanao and more distal islands are equal to or greater than the 
indices between Palawan and these same islands, less isolation of island faunas 
along eastern routes from Mindanao would be indicated. 

All indices, those for each of the taxa, (frogs, lizards, and snakes), as well 
as those for the total herpetofaunas (tables 6-9), are moderately high when 



Vol. XXXVIII] BROWX AXD ALCALA : HERPETOFAVNA OF PHILIPPINES 121 

Table 8. Indices of similarity for snakes. (The index in parentheses is based on species 
exclusive of the 23 widespread forms associated with man's economy or beach communities, 
that in brackets is based on the species remaining after r elicit species are also excluded). 





Mindanao 


Leyte 


Bohol 


Negros 


Mindoro 


Luzon 


Palawan 


0.528 


0.431 


0.491 


0.323 


0.308 


0.308 




(0.373) 


(0.229) 


(0.278) 


(0.182) 


(0.171) 


(0.296) 




[0.385] 


[0.235] 


[0.286] 


[0.190] 


[0.176] 


[0.314] 


Mindanao 




0.596 


0.644 


0.500 


0.414 


0.538 






(0.488) 


(0.524) 


(0.360) 


(0.293) 


(0.433) 






[0.474] 


[0.513] 


[0.348] 


[0.268] 


[0.436] 


Leyte 






0.684 


0.468 


0.324 


0.491 








(0.609) 


(0.305) 


(0.182) 


(0.341) 








[0.571] 


[0.345] 


[0.200] 


[0.378] 


Bohol 








0.531 
(0.315) 
[0.345] 


0.462 
(0.348) 
[0.381] 


0.542 
(0.333) 

[0.378] 


Negros 










0.542 
(0.516) 
[0.429] 


0.706 
(0.640) 
[0.622] 


Mindoro 












0.621 
(0.585) 
[0.541] 



Palawan and Mindanao are compared, even though the differences in diversity 
are relatively large. This is the result of the large number of species which these 
2 entry- way islands have in common. These common species have either entered 
the 2 islands relatively recently from Borneo or have continued to reinvade from 
time to time. The fact that many of these species have not dispersed to more 
distal islands in the Philippines and yet are consiDecific with Bornean poiDulations 
tends, however, to support the first explanation of more recent entry. 

When indices between Mindanao and IVIindoro, Leyte, Luzon, or Negros, 
based on the complete fauna (both native and nonnative species), are compared 
with the indices between Palawan and these same islands (tables 6-9), those with 
Mindanao are much higher, with one excei^tion, even though the greater diversity 
of the Mindanao fauna, almost twice that of Palawan, tends to depress the 
similarity indices. The exception is the Palawan-Mindoro index for lizards. 
These differences are more jDronounced and the excejDtion ceases to exist when 
the 23 widespread, nonnative sjaecies (p. Ill) are excluded. The differences are 
not quite as great, in most instances, when those classified as older relicts (p. 117) 
are also excluded; but these latter, small differences can be accounted for by 
the absence of the relict genera Platvmantis and Brachvmelcs from Palawan. 



122 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

Table 9. Indices of similarity for the total herpetofauna. {The index in parentheses is 
based on species exclusive of the 23 widespread forms associated with man's economy or 
beach communities, that in brackets is based on the species remaining after relicit species 
are also excluded). 





Mindanao 


Bo /to I 


Negros 


Mindoro 


Luzon 


Palawan 


0.453 


0.453 


0.372 


0.382 


0.372 




(0.304) 


(0.245) 


(0.109) 


(0.165) 


(0.186) 




[0.331] 


[0.274] 


[0.122] 


[0.188] 


[0.214] 


Mindanao 




0.660 


0.493 


0.415 


0.475 






(0.584) 


(0.354) 


(0.286) 


(0.362) 






[0.567] 


[0.337] 


[0.258] 


[0.331] 


Bohol 






0.547 


0.508 


0.494 








(0.377) 


(0.364) 


(0.368) 








[0.345] 


[0.342] 


[0.337] 


Negros 








0.588 
(0.457) 
[0.415] 


0.616 
(0.511) 
[0.481] 


Mindoro 










0.605 
(0.523) 
[0.505] 



This indicates that the primary dispersal routes for the native fauna were by 
way of Mindanao, and that the present herpetofauna of Palawan has not dis- 
persed widely into other areas in the Philippines. 

The high indices between Mindanao, Leyte, and Bohol are due to the very 
large number of species which they hold in common, even though diversity is 
greatly reduced. This suggests a very high rate of exchange between Mindanao 
and these 2 nearby islands on the eastern routes. The slightly higher indices 
between Mindanao and Bohol as compared to those between Mindanao and 
Leyte may be partly the result of lesser knowledge of the fauna of Leyte. How- 
ever, it may also be the result of some exchange across a direct Mindanao- 
Bohol route. 

The relatively high Mindoro-Negros index for each taxon, as well as for the 
total herpetofauna, may be in part due to their positions at the distal end of 
dispersal routes, and, consequently, the same species have tended to reach both. 
However, it also suggests an active Mindoro-Panay-Negros dispersal route. The 
richer fauna of Negros, on the other hand, also indicates that a part of the 
Negros fauna must have arrived by way of the shorter Mlndanao-Leyte-Bohol, 
or Mindanao-Ley te-Cebu dispersal routes, or in some instances perhaps, a 
Mindanao-Bohol-Negros route. 

The progressively lower indices with Mindanao, as one progresses along the 



Vol. XXXVIII] BROWN AND ALCALA: HERPETOFAUNA OF PHILIPPINES 123 

eastern dispersal route to Mindoro, appears to be consistent with the decrease 
in diversity. The Mindoro-ISIindanao index is higher for frogs than for either 
lizards or snakes. This would be expected in terms of a later arrival, which might 
be accounted for by their slower dispersal ability where marine barriers have 
presumably operated. 

Evidence for Over-Water Dispersal 

In addition to the evidence of barriers to dispersal indicated by the reduction 
of numbers of species on distal islands as compared to proximal islands, when 
the size variable is minimized, as suggested in the section on diversity (fig. 2), 
the data may be further analyzed in terms of the probable differential effect of 
marine barriers. As noted at the close of the previous section, on biological 
grounds marine barriers would be expected to affect adversely dispersal of 
amphibians to a greater extent than that of reptiles. Therefore, on the assumption 
that the primary entryway into the Philippine archipelago was from Borneo by 
way of INIindanao, the amphibia might be expected to exhibit a more rapid 
reduction in number of species than do the reptiles when comparisons are made 
between Mindanao and the more distal islands such as Bohol, Negros, Mindoro, 
and Luzon if dispersal did take place wholly or in part across such barriers along 
these eastern routes. This should become even more evident if, as well as the 
nonnative, the presumably older endemic relict elements, those which exhibit the 
typical relict pattern of chance pockets of isolated species and or secondary 
radiation centers on distal islands, were also excluded. 

To evaluate this, we propose a simple proportional-diversity index. This 
makes possible an objective comparison of changes in diversity for amphibia 
relative to changes in diversity for the reptiles. The index is calculated as the 
ratio of the number of species in the particular taxon (frogs, lizards, etc.) to 
the number of species in the total herpetofauna. 

This index for frogs is indeed lower for Luzon, Negros, and INIindoro, at the 
distal end, than for Mindanao or Bohol at the proximal end of the dispersal 
routes. The amphibian index for Bohol is noteworthy in that it is slightly 
higher than that for Mindanao. This may be, at least in part, a distortion due 
to a disproportionately poorly known snake fauna (see p. 119). 

Since indices for reptiles (both lizards and snakes) tend to increase as the 
index for amphibians decreases, it is interesting to note that the index for lizards 
exhibits a greater increase than that for snakes in all instances, except for Luzon 
Island, when the distal islands are compared with Mindanao at the entryway. 
For some reason, snakes appear to have been relatively more successful in their 
dispersal to Luzon, or have suffered fewer extinctions there. 

We interpret these changes in the proportional diversity indices as supporting 
the conclusion that much of the herpetofauna of the Philippines, with the ex- 
ception of that of Palawan, has been the result of waif dispersal across marine 



124 



CALIFORNIA ACADEMY OF SCIENCES 



[Proc. 4th Ser. 



JIM 1 


' 


1 1 1 1 1 1 


I 1 1 


_ 


- 








: 


- 








- 








C/5 

< 




; 


.. ^" 




m 


z 


- 


^ \ 




X 


' 


- 


\ 


y 


Q. 


_ 


- 


\ 


\ 


^ 


- 


- 


\ 


\ 


< 


- 


- 




.\ 




- 


— 


\ 


\r- 




— 


I 


\ 
\ 


•" 




; 


- 


\ 


V 




- 


1 1 1 1 1 


1 1 1 


1 


1 1 1 


- 



_ 










: 


" 


i\ 






to 

UJ 

< 

-z. 

CO 


- 


- 


1 \ 


\ 






- 


- 


1 
1 


\ 


\ 




- 


: 






\. 




: 


- 




I 


•' 




- 


1 1 1 1 


1 1 


1 


1 1 1 1 


1 1 1 1 1 


- 



cq 



S3i33dS do yaawnN 



S3i33ds 30 aaawnN 





1 1 1 1 1 


1 1 1 1 1 1 1 


i 


- 


< 




- 


- 


2 

3 




- 




^ 
^ 






.. . 


UJ 


'^'\ \ 


— 


- 


CL 




- 




cr 


\ \ 


~ 


_ 


LU 


\ \ 


_ 


- 


X 


\ \ 


- 


- 


_l 
< 


\ V 


- 


- 


H- 




_ 




o 






— 


1- 




—_ 


- 






- 


1 1 1 


1 1 1 1 1 


1 1 1 1 I 1 1 1 


1 



I I I I I I — I — I r 



CO 

Q 

cr 
< 

M 




1 1 I I I I I i_ 



8 < 



S3Q3dS 30 M39WnN 



S3l33dS 30 d3awnN 



P .s 



Vol. XXXVIII] BROWN AND ALCALA: HERPETOFAUNA OF PHILIPPINES 125 

Table 10. Proportional diversity indices for caecilians, frogs, lizards, and snakes for 
selected proximal, intermediate, and distal islands for the total herpetofauna, the fauna with 
the presumed nonnative species excluded {in parentheses), and with both the nonnative and 
that element classified as relict included [in brackets]. 





Palawan 


Mindanao 


Bo hoi 


Ncgros 


Mindoio 


Luzon 


Number of Caecilian species 


0.013 


0.008 










(Total herpetofaunal species) 


(0.018) 
[0.019] 


(0.009) 
[0.011] 










Number of frog species 


0.269 


0.272 


0.292 


0.205 


0.207 


0.223 


(Total herpetofaunal species) 


(0.309) 


(0.291) 


(0.352) 


(0.218) 


(0.216) 


(0.230) 




[0.302] 


[0.292] 


[0.333] 


[0.178] 


[0.188] 


[0.186] 


Number of lizard species 


0.295 


0.416 


0.431 


0.423 


0.466 


0.362 


(Total herpetofaunal species) 


(0.218) 


(0.408) 


(0.392) 


(0.400) 


(0.459) 


(0.351) 




[0.208] 


10.391] 


[0.381] 


[0.400] 


[0.469] 


[0.322] 


Number of snake species 


0.423 


0.312 


0.278 


0.372 


0.328 


0.415 


(Total herpetofaunal species) 


(0.455) 


(0.301) 


(0.255) 


(0.382) 


(0.324) 


(0.419) 




[0.422] 


[0.315] 


[0.286] 


[0.422] 


[0.344] 


[0.492] 



barriers from Mindanao Island. This conclusion is most strongly evidenced 
when nonnative and isolated relict sfDecies are excluded leaving the so-called 
immigrant species (table 10). 

Diversity as Related to Area and Distance 

Area. When number of native species is plotted against area (fig. 6) for the 
compact group of islands included in the present study, the curves for amphibians 
and the total herpetofauna exhibit the expected pattern for overwater dispersal 
as postulated by MacArthur and Wilson (1963, 1967). Large islands do exhibit 
a greater diversity (number of species) than small islands and near islands a 
consistently greater diversity than more distant islands, in terms of the probable 
migration routes. 

The diversity of lizards on Palawan, a near small island, and the diversity 
of snakes of Mindanao are somewhat less than expected, however, based on the 
slope of the curves. The diversity of snakes on Luzon, the most distant large 
island, also appears slightly greater than might be expected. One explanation 
might be that the relatively short over-water distances obtaining in this compact 
archipelago have made possible a higher frequency of invasion of snakes along 
the eastern chain to Luzon or that there are more relict snakes on Luzon. Since 
lizards of the endemic genus Brachymeles are absent from Palawan, this probably 
accounts in part for the lower diversity of this faunal group on that island. The 
diversity for amphibians and lizards on Bohol Island is greater than expected 
and that of the snakes lower. It has been suggested that the latter may be due 



126 



CALIFORNIA ACADEMY OF SCIENCES 



[Proc. 4th Ser. 



JIM 1 


1 1 1 




1 M 1 1 1 


1 1 


- 


- 








(f) 


- 


~ 




/ 


/ 


< 


" 




, / 


/ 
/ 


/ 






— 


^/ 




1 


S 


— 


- 


-J/ 
/ 


""/ 




< 


- 


- 


/ 


~7 






- 


- 


1 
/* 








- 


1 1 1 1 1 


1 1 1 


/- 


1 1 1 1 1 


1 1 


o 




S3i03ds JO asawnN 



S3io3ds 30 aaawnN 



J 1 


1 1 1 1 




1 1 


1 1 


_ 


- 


< 








- 


- 


2 
3 








- 


- 


< 






/ 


_ 




U_ 








~ 


O 

1- 
UJ 
Q. 


/ 


• Ul 


[ 


~ 




QL 


; 








- 


UJ 


i^l 






- 


_ 


X 


SI 


■jj 




_ 






1/ 








_ 


—I 


/ 


5/ 






' 


< 

1- 


/ 
1 






- 


_ 


O 


1 








1 1 


1- 
1 1 1 1 1 


1 

1 

1 II 


1 1 


1 1 





II I I — I — I — I r 



S303dS 30 d3awnN 




I I I I I I 



S3ID3dS 30 a3BwnN 



« o 

O 3 



vO 



Vol. XXXVIII] BROWN AXD ALCALA : HERPETOFAVXA OF PHILIPPINES 127 

Table U. Calculated distance index for Philippine islands included in this study in rela- 
tion to Borneo. 

Weighted number Approximate over-water Weighted 

of effective distance by present distance 

Route tnarine barriers probable routes (in km.) index 

Borneo— Palawan (1) ISO ISO 

Borneo — Sulus — Mindanao (1) 345 345 

Borneo— Mindanao— Leyte (2) 395 790 

Borneo — Mindanao— Leyte — Bohol (3) 420 1260 
Borneo — Mindanao — Leyte — Cebu 

— Negros (4-5=41/2) 505 2272.5 
Borneo — Mindanao — Leyte — Samar 

—Luzon (4) 440 1760 
Borneo — Mindanao — Leyte — Samar 

— Luzon — Mindoro (5) 455 2275 



to a sampling bias. The high diversity of lizards and amphibians may possibly 
be the result of the very narrow water gap between Leyte and Bohol. 

Distance. Any attempt to measure over-water dispersal distances in this 
compact, nonlinear archipelago is difficult. If the effective distance is measured 
as the airline distance from Borneo to the various islands, Leyte is almost as 
distant as Luzon, and Bohol almost as distant as Mindoro (fig. lb). If the 
effective distance is measured as the sum of the breadth of over-water distances 
between islands, by ways of eastern migration routes for all islands except 
Palawan, assuming that the islands themselves provide stepping stones of 
ecologically relatively uniform space, the effective over-water distance between 
Mindanao and Luzon, the northernmost island, is 75 kilometers as compared to 
55 to 90 kilometers as the effective over-water distance between Mindanao and 
Bohol for example. In an attempt to minimize these sources of error we have 
derived a weighted index by multiplying the sum of the approximate over-water 
distances times a value for number of marine barriers (table 11). 

When this distance index is plotted against number of species (fig. 7), the 
shape of the curves are again, with slope opposite to that of the area curves, in 
keeping with that expected for amphibians and the herpetofauna as a whole, 
and the curves for large and small islands are nearly parallel. The diversities 
for snakes on Mindanao and Luzon and for lizards on Palawan also impose 
effects on the slopes of the curves which are comparable to the effects on the 
area curves. Bohol and Leyte also exhibit a very low diversity for snakes 
relative to the curve for other islands in the same general size-category. The 
general agreement between these curves and those based on area suggests that 
such a weighted distance index may be useful in the island faunas in similar 
compact archipelagos. 



128 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

SUMMARY AND CONCLUSIONS 

The existence of relict patterns and secondar}' radiation centers, the relative 
importance of possible internal migration routes from alternative entryways, 
the evidence for over-water dispersal of the "migrant" element, and the effects 
of island area and distance are considered for the herpetofauna of the Philippines, 
a compact, fringing archipelago. Evaluations are based primarily on diversities 
and relationships of the herpetofaunas of 7 of the islands, which have a total of 
220 species (54 amphibians, 92 hzards, and 74 snakes). Only in the evaluation 
of relict distributions is the known herpetofauna of the total archipelago taken 
into consideration. 

The distributional patterns within the Philippines of the endemic, multi- 
species genera of lizards, Luperosaurus and Brachymeles, as well as the amphibian 
genus Platymantis, exhibit relict patterns of the type resulting from partial ex- 
tinction of an old fauna, which has existed as a number of isolated units. 
Brachymeles and Platymantis also give evidence of secondary radiation centers 
in distal islands. Patterns for the endemic genus of lizards Pseudogekko and 
snake genera Cyclocorus, Hologcrrhum, and Oxyrhabdium are simpler relict 
patterns with only 1 or 2 species in each genus. These are rather widely distrib- 
uted, or, in some instances, limited to either the distal or proximal islands. 
These patterns are those postulated by Darlington (1957) as patterns which 
would develop within a chain of islands. 

Sorenson's index of similarity is used to evaluate the relative effectiveness 
of dispersal routes. These indices, particularly when the nonnative fauna is 
excluded, indicate that the primary dispersal route or routes within the archi- 
pelago have been the eastern routes, by way of the Mindanao-Leyte — or possibly, 
in some instances the Mindanao-Bohol — pathways. The Palawan entryway has 
contributed very little to the herpetofauna of the rest of the Philippines. High 
indices of similarity between Negros and Mindoro suggest a relatively active 
migration route between these two islands. 

A proportional-diversity index, based on the presumed lower ability of am- 
phibians to disperse across marine barriers, is used to evaluate the probable 
effect of marine barriers. The evidence indicates that, with the exception of 
Palawan Island, much of the herpetofauna has apparently reached the inter- 
mediate and distal islands of the archipelago from Mindanao as a result of waif 
dispersal across marine barriers. 

When number of species for the total herpetofauna, and for the amphibians, 
lizards, and snakes independently, are plotted against area or against a weighted 
distance value the curves for amphibia and the total herpetofauna exhibit 
patterns consistent with those projected from MacArthur's and Wilson's thesis 
(1967) regarding faunal diversity along a chain of islands of varying size. The 
data indicate, however, that the diversity of the lizard fauna on Palawan island 



Vol. XXXVIII] BROWN AND ALC ALA : HERPETOFAUNA OF PHILIPPINES 129 

is lower than expected, and that the diversity of the snake fauna is probably 
somewhat lower for Mindanao and higher for Luzon island relative to the 
diversity exhibited by the fauna of other islands in the study. The effects of 
the narrow marine barriers on the present distribution of the amphibian fauna 
have produced a pattern of island diversities in general agreement with the 
MacArthur-Wilson hypothesis. The diversities of lizards and snakes for the 
sample group of islands included in this study exhibit several discrepancies. The 
relatively narrow over-water barriers between the islands of this compact archi- 
pelago, less effectual against reptiles than amphibians, and the several possible 
migration routes and secondary centers of radiation may be factors in the 
distribution patterns of these faunal elements. The reasons, however, for the 
low diversity of lizards of Palawan and for the effectiveness of the barrier be- 
tween Palawan and Mindoro, a marine channel only about 150 km. in breadth 
at the present time and broken by small islands, are not readily explained from 
data. 

ACKNOWLEDGMENTS 

This study was supported by National Science Foundation Grant GB-4156. 
Illustrations were prepared by Walter Zawojski, Stanford University. 

LITERATURE CITED 

Brown, W. C, and A. C. Alcala 

1961. Populations of amphibians and reptiles in the submontane and montane forests 
of Cuernos de Negros. Ecology, vol. 42, pp. 628-636. 

1964. Relationship of the herpetofaunas of the non-dipterocarp communities to that of 

the dipterocarp forest of southern Negros Islands, Philippines. Senkenbergiana 
Biologica, Frankfurt am Main. vol. 45, pp. 519-611. 
Brown, W. C, and D. S. Rabor 

1967. Review of the genus Brachymeles (Scincidae), with descriptions of new species 
and subspecies. Proceedings of the California Academy of Sciences, vol. 34, 
no. 16, pp. 525-548. 
Darlington, P. J., Jr. 

1957. Zoogeography: the geographical distribution of animals. Wiley and Sons, New 
York, xi + 675 pp. 
Inger, R. F. 

1954. Systematics and zoogeography of Philippine amphibia. Fieldiana: Zoology, vol. 
33, pp. 183-531. 
Inger, R. F., and H. Marx 

1965. The systematics and evolution of the oriental colubrid snakes of the genus 

Calamaria. Fieldiana. vol. 49, pp. 1-304. 
Leviton, a. E. 

1963. Remarks on the zoogeography of Philippine terrestrial snakes. Proceedings of the 
California Academy of Sciences, vol. 31, no. 15, pp. 369-416. 
MacArthur, R. H., and E. B. Wllson 

1963. .\r\ equilibrium theory of insular zoogeography. Evolution, vol. 17, pp. 373-387. 
1967. The theory of island biogeography. Princeton University Press, xi + 203. 



130 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

SORENSON, T. 

1948. A method of establishing groups of equal amphtude in plant sociology based on 

similarity of species content and its apphcation to analyses of the vegetation 

on Danish commons. Danske Videnskabernes Selskab, Biologiske Skrifter, 

Copenhagen, vol. S, no. 4, pp. 1-34, 6 tables, 1 figure. 
Taylor, E. H. 

1922a. The lizards of the Philippine Islands. Philippine Bureau of Science. Manila. 

Publication 17, 269 pp., 23 p]s. 
1922b. Additions to the herpetological fauna of the Philippine Islands I. Philippine 

Journal of Science, vol. 21, pp. 161-206, 7 pis. 
192 2c. Additions to the herpetological fauna of the Philippine Islands II. Philippine 

Journal of Science, vol. 21, pp. 257-303, 4 pis. 
1923. Additions to the herpetological fauna of the Philippine Islands III. Philippine 

Journal of Science, vol. 22, pp. 515-557, 3 pis. 
1925. Additions to the herpetological fauna of the Philippine Isalnds IV. Philippine 

Journal of Science, vol. 26, pp. 97-111. 
1928. Amphibians, lizards and snakes of the Philippines. In: Dickerson, R., et al. 

1928, pp. 214-241, pis. 27-32. 

ZWEIFEL, R. G. 

1967. Identity of the frog Cornujer unicolor and application of the name Cornufer. 
Copeia, 1967, pp. 117-121. 



PROCEEDINGS 

OF THE 

CALIFORNIA ACADEMY OF SCIENCES 

FOURTH SERIES 



Festschrift for George Sprague Myers 



Vol. XXXVIII, No. 7, pp. 131-138. December 31, 1970 



TROPICAL SHELF ZOOGEOGRAPHY 

By 

John C. Briggs 
Department of Zoology, University of South Florida, Tampa, Florida 

INTRODUCTION 

The richest marine fauna is found in the shallow waters of the tropical 
oceans at depths generally less than 200 meters. Zoogeographically, four 
great regions may be identified, the Indo-West Pacific, the Eastern Pacific, 
the Western Altantic, and the Eastern Atlantic. Each region may, in turn, be 
subdivided into provinces, but these will not be discussed at this time. To the 
north and south, the tropics are bounded by the 20°C. isotherm for the coldest 
month in the year. Longitudinally, the tropical regions are separated from one 
another by barriers that are very effective since each region possesses, at the 
species level, a fauna that is highly endemic. By studying the operation of 
these longitudinal barriers, one can learn something about the interrelationship 
of the regions and can also obtain information leading to a better understand- 
ing of zoogeography and evolution. 

THE EAST PACIFIC BARRIER 

The Indo-West Pacific and Eastern Pacific regions are separated by the 
East Pacific Barrier, the vast stretch of deep-water that Ues between Polynesia 
and America. In regard to the shore fishes, it was concluded that an eastward 
colonization movement was taking place across the Barrier and that successful 
reciprocal migrations were, at least, very rare and might be completely lacking 

[131] 



132 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

(Briggs, 1961, 1964, 1966). When the general relationship of the tropical 
shelf regions was first discussed (Briggs, 1967a), comparable information on 
the major groups of the shallow-water invertebrates was not available. 

Emerson (1967) pubHshed a revealing analysis of the distribution of 
those Indo-West Pacific species of mollusks that have succeeded in penetrating 
across the Barrier into the Eastern Pacific. He found that such trans-Pacific 
species were largely restricted, in the Eastern Pacific, to the oceanic islands, 
the greatest numbers being found at CUpperton (33 species) and at the 
Galapagos (25 species). It was noted that the gastropods, which greatly out- 
numbered the bivalves, belonged to groups that were known to have relatively 
long larval stages. Most important of all, Emerson pointed out that no mol- 
luscan species of apparent Eastern Pacific origin were known to occur in 
Polynesia. 

Data on the other invertebrate groups are not as complete, but it is 
significant that some of the Uttoral echinoderms (Ekman, 1946), holothurians 
(Deichmann, 1963), decapod crustaceans (Chace, 1962; Garth, 1965), and 
hermatypic corals (Emerson, 1967) found in the Eastern Pacific (especially 
around the offshore islands) are trans-Pacific species of apparent Indo-West 
Pacific origin. Therefore, it may now be said that for the tropical marine 
shore fauna in general, including both fishes and invertebrates, it seems Ukely 
that successful migration across the East Pacific Barrier takes place in one 
direction only — from west to east. 

THE NEW WORLD LAND BARRIER 

The New World Land Barrier, with the Isthmus of Panama forming its 
narrowest part, is virtually a complete block to the movement of tropical 
marine species between the Eastern Pacific and Western Atlantic. This state 
of affairs has existed since about the latest Pliocene or earhest Pleistocene 
(Simpson, 1965; Patterson and Pascual, 1968) so that, at the species level, 
the two faunas are well separated. The present Panama Canal has not 
notably altered this relationship since, for most of its length, it is a freshwater 
passage forming an effective barrier for all but a few euryhaline species. 

The New World Land Barrier is the most effective of the four zoogeo- 
graphic barriers that separate the tropical faunal regions. It has stood for 
approximately three million years, but it now appears that man is about to 
breach this barrier by excavating a sea-level canal somewhere in the vicinity of 
the Isthmus of Panama. If such a canal is constructed, it would present ample 
opportunities for marine animals to migrate in either direction. This could 
result in the Eastern Pacific being invaded by over 6000 species of fishes and 
invertebrates and the Western Atlantic being invaded by over 4000 species. 
Since the Western Atlantic species would apparently be competitively dominant, 
it has been predicted that a large scale extinction would take place in the 



Vol. XXXVIII] BRIGGS. TROPICAL SHELF ZOOGEOGRAPHY 133 

Eastern Pacific resulting in the irrevocable loss of a huge number (possibly 
thousands) of species (Briggs, 1968, 1969). 

THE MID-ATLANTIC BARRIER 

The broad deep-water barrier that separates the Western Atlantic tropics 
from those of the West African coast functions in a very interesting manner. 
An impressive number of shore fishes have managed to traverse the Mid- 
Atlantic Barrier from west to east. It has been estimated (Briggs, 1967a) that 
about 118 shore fish species have trans-Atlantic distributions but that only 
about 24 of them came from the Indo-West Pacific via the Cape of Good Hope. 
The rest probably evolved in the Western Atlantic and successfully performed 
an eastward colonization journey across the ocean. Trans-Atlantic species 
comprise about 30 percent of the shore fish fauna of tropical West Africa. 

Works on some of the major groups of West African invertebrates also 
show that an appreciable number of the species are trans-Atlantic: Dekeyser 
(1961) found that about 25 percent of the ascidians showed this distribution; 
Burton (1956), 18 percent of the sponges; Monod (1956), 16 percent of the 
anomuran and brachyuran crabs; Knudsen (1956), 6 percent of the prosobranch 
mollusks; Ekman (1953), 16 percent of the starfishes, brittle stars, and sea 
urchins; and Marcus and Marcus (1966), 29 percent of the opisthobranch 
mollusks. Furthermore, Chesher (1966), who found 8 trans-Atlantic species of 
sea urchins in the Gulf of Guinea, stated that gene flow appeared to take place 
from west to east. 

It seems apparent that, in both the fishes and the invertebrates, the great 
majority of the trans-Atlantic species originated in the Western Atlantic and 
then migrated eastward. The westward colonization traffic appears to be re- 
stricted to certain dominant species that originated in the Indo-West Pacific 
and then gained access to the Atlantic by rounding the Cape of Good Hope. 
So far, there are no indications that species originating in the Eastern Atlantic, 
and belonging to genera typical of that area, have been successful in becoming 
established on the western side. 

THE OLD WORLD LAND BARRIER 

The Eastern Atlantic and the Indo-West Pacific regions are separated 
by the Old World Land Barrier. It has been estimated that the continental 
masses of Eurasia and Africa have been linked at least since the beginning 
of the Pleistocene (Gohar, 1954). The Suez Canal is a sea-level passage that 
has been open since 1869 but migration of marine animals has been inhibited 
for two reasons: first, the canal connects two areas that are separated by a 
temperature barrier, the Red Sea being tropical while the Mediterranean is 
warm-temperate; second, the Bitter Lakes, which form part of the Suez pas- 
sageway, have a high salinity (about 45 percent). Despite these difficulties, 



134 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

the limited migratory movements that have taken place through the Suez 
Canal do provide some significant information. 

The Mediterranean has been invaded by at least 24 species of Red Sea 
fishes (Ben-Tuvia, 1966), 16 species of decapod crustaceans (Holthuis and 
Gottlieb, 1958), and several species in other groups such as the tunicates 
(Peres, 1958), mollusks (Engel and van Eeken, 1962), and stomatopod crus- 
taceans (Ingle, 1963). So, while there is ample evidence of intrusions into 
the eastern Mediterranean, there are no reliable data that indicate any success- 
ful reciprocal migration. Also, there are some indications that the invaders 
from the Red Sea (a part of the Indo-West Pacific Region) are replacing 
rather than coexisting with certain native species (George, 1966). 

The various circumtropical shore species have probably been able to pre- 
serve their genetic homogeneity by means of migration around the Cape of 
Good Hope (in addition to crossing the open ocean barriers in the Pacific 
and Atlantic). Talbot and Penrith (1962) remarked that surface temperatures 
of 21°C. are often present round the Cape outside a cold upwelling area. There 
are about 16 known species of circumtropical shore fishes (Briggs, 1960). 
Besides these, there are about 15 other species that apparently transgress 
the Old World Land Barrier at the Cape (Briggs, 1967a). Of the total of 31 
fish species, eight are monotypic but all the rest represent genera that are 
best developed in the Indo-West Pacific. 

Apparently, only a few tropical invertebrate species have been able to 
migrate around the Cape of Good Hope. Monod (1956) in his monographic 
study of the West African decapods showed that 10 out of 176 shore species 
occurred in the Indo-West Pacific. Ekman (1953) noted that only 2 percent 
of the tropical Atlantic echinoderms (Asteroidea, Ophiuroidea, and Echinoidea) 
extended around the Cape. It appears, especially from the ichthyological evidence, 
that the colonization movement of tropical shore species around the Cape takes 
place entirely in a westerly direction, from the Indo-West Pacific into the Atlantic. 

RELATIONSHIPS OF THE SHELF REGIONS 

Evidence now available about the dispersal of the shallow-water marine 
invertebrates tends to substantiate the general nature of a remarkable distri- 
butional phenomenon that was discovered earlier for the shore fishes (Briggs 
1961, 1964, 1967a). Successful (colonizing) migrations across the zoogeo- 
graphic boundaries that delimit the Indo-\A^est Pacific can apparently take 
place in one direction only, outward into areas where the fauna is poorer and 
the competition is less. The realization that the East Pacific and Old World 
Land Barriers operate as one-way filters enables us to understand better how 
the Indo-West Pacific Region serves as the evolutionary and distributional 
center for the tropical shore animals of the world. We can see that competi- 
tively dominant species continue to migrate, as they probably have for mil- 



Vol. XXXVIII] BRIGGS: TROPICAL SHELF ZOOGEOGRAPHY 135 

lions of years, from the Indo-West Pacific eastward across the open ocean to 
America and westward around the Cape of Good Hope into the Atlantic; 
since 1869, some of them have also been able to pass northward through the 
Suez Canal into the Mediterranean. 

The Western Atlantic Region may be considered a secondary center of 
evolutionary radiation. Many species evolved in this area have proved ca- 
pable of migrating eastward to colonize the tropical Eastern Atlantic. How- 
ever, species originating in the Eastern Atlantic are apparently incapable of 
successfully invading the western side. Again, the advantage seems to lie 
with the area that possesses the richer fauna and the higher level of competi- 
tion. 

It can be seen that the completely eastward direction of successful mi- 
gratory movements across the East Pacific Barrier and the predominantly 
eastward movements across the Mid-Atlantic Barrier take place in a direction 
opposite to that of the main flow of the surface waters via the North and 
South Equatorial Currents. In contrast, the surface and subsurface counter- 
currents in the tropical Pacific and Atlantic are weakly developed but these 
smaller currents are obviously the principal means by which successful trans- 
port is achieved. 

Fell (1967) noted that certain groups of shore species apparently demon- 
strated speciation gradients in which the number of species gradually dimin- 
ished around the world in a westward direction. He interpreted this to mean 
that the direction of successful migrations had also been to the west and that 
such dispersals had been carried out by the North and South Equatorial cur- 
rents. Subsequently, it was pointed out that the existence of a gradient in 
numbers of species (or genera) across a major barrier did not necessarily 
indicate the direction of the original successful migration (Briggs, 1967b). 
The fact that colonizations do take place in a direction opposite to that of 
the major currents is a good indication that biological competition rather 
than passive transport is probably the most important factor controlling the 
successful dispersal of tropical marine shore animals. 

LITERATURE CITED 

Ben-Tuvia, a. 

1966. Red Sea fishes recently found in the Mediterranean. Copeia, no. 2, pp. 254-275, 
2 figs. 
Briggs, J. C. 

1960. Fishes of worldwide (circumtropical) distribution. Copeia, no. 3, pp. 171-180. 

1961. The East Pacific Barrier and the distribution of marine shore fishes. Evolution, 

vol. IS, no. 4, pp. 545-554, 3 figs. 
1964. Additional transpacific shore fishes. Copeia, no. 4, pp. 706-708. 
1966. Zoogeography and evolution. Evolution, vol. 20, no. 3, pp. 282-289. 
1967a. Relationship of the tropical shelf regions. Studies in Tropical Oceanography, 

Miami, no. 5, pp. 569-578. 



136 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

1967b. Dispersal of tropical marine shore animals: coriolis parameters or competi- 
tion? Nature, vol. 216, no. 5113, page ,350. 

1968. Panama's sea level canal. Science, vol. 169, no. 3853, pp. 511-513. 

1969. The sea-level canal: potential biological catastrophe. BioScience, vol. 19, no. 1, 

pp. 44-47. 
Burton, M. 

1956. The sponges of West Africa. Atlantide Report, no. 4, pp. 111-147, 4 figs. 
Chace, F. 

1962. The non-brachyuran decapod crustaceans of Clipperton Island. Proceedings of 

the United States National Museum, vol. 113, no. 3466, pp. 605-635, 7 figs. 
Chesher, R. H. 

1966. Report on the Echinoidea collected by R/V Pillsbury in the Gulf of Guinea. 

Studies in Tropical Oceanography, Miami, no. 4, part 1, pp. 209-223. 
Deichmann, E. 

1963. The holothurians of Clipperton Island in the Eastern Tropical Pacific. Breviora, 

no. 179, pp. 1-5. 
Dekeyser, p. L. 

1961. Liste provisoire des urocordes de la cote occidentale d'Afrique. Bulletin de la 

Institute fran(;aise d'Afrique Noire, serie A, vol. 23, no. 1, pp. 217-230. 
Ekman, S. 

1953. Zoogeography of the sea. Sidgwick and Jackson, London, xiv + 417 pages, 

121 figs. 
Emerson, W. K. 

1967. Indo-Pacific faunal elements in the tropical eastern Pacific, with special refer- 

ence to the mollusks. Venus, Japanese Journal of Malacology, vol. 25, nos. 
3 and 4, pp. 85-93. 
Engel, H. and C. ]. van Eeken 

1962. Red Sea Opisthobranchia from the coast of Israel and Sinai. Sea Fisheries Re- 

search Station, Haifa, Israel, Bulletin, no. 30, pp. 15-34, 7 figs. 
Fell, H. B. 

1967. Resolution of coriolis parameters for former epochs. Nature, no. 214, pp. 1192- 

1198. 
Garth, John S. 

1965. The brachyuran decapod crustaceans of Clipperton Island. Proceedings of the 

California Academy of Sciences, vol. 33, no. 1, pp. 1-46, 26 figs. 
George, C. J. 

1966. A two year study of the fishes of the sandy littoral of St. George Bay, Leb- 

anon. Abstracts, 2nd International Oceanographic Congress, Moscow, page 130. 
Gohar, H. A. F. 

1954. The place of the Red Sea between the Indian Ocean and the Mediterranean. 

Publications of the Hydrobiological Research Institute, University of Istanbul, 

Series B, vol. 2, no. 2-3, pp. 1-38, map. 
HoLTHUis, L. B. and E. Gottlieb 

1958. An annotated list of the decapod Crustacea of the Mediterranean coast of 

Israel, with an appendix listing the decapoda of the eastern Mediterranean. 

Bulletin of the Sea Fisheries Research Station, Israel, no. 18, pp. 1-126, 15 

figs. 
Ingle, R. W. 

1963. Crustacea Stomatopoda from the Red Sea and the Gulf of Aden. Bulletin of 

the Sea Fisheries Research Station, Haifa, Israel, no. 33, pp. 1-69, 73 figs. 



Vol. XXXVIII] BRIGGS: TROPICAL SHELF ZOOGEOGRAPHY 137 

Knudsen, Jorgen 

1956. Marine prosobranchs of tropical West Africa (Stenoglossa). Atlantide Report, 
no. 4, pp. 8-110, 2 figs., 4 pis. 
Marcus, Ernst, and Eveline Marcus 

1966. Opisthobranchs from tropical West Africa. Studies in Tropical Oceanography, 
Miami, no. 4, part 1, pp. 152-208, 62 figs. 
MONOD, Th. 

1956. Hippidea et brachyura ouest-africains. Memoires de la Instiute francaise 
d'Afrique Noire, no. 45, pp. 1-674, 884 figs., 1 map. 
Patterson, B. and R. Pascual 

1968. The fossil mammal fauna of South America. The Quarterly Review of Biology, 
vol. 43, no. 4, pp. 409-451, 13 figs. 
Peres, J. M. 

1958. Ascidies recoltees sur les cotes Mediterraneenar d'lsrael. Bulletin of the Sea 
Fisheries Research Station, Haifa, Israel, no. 19, pp. 143-150. 
Simpson, G. G. 

1965. The geography of evolution. Chilton Books, Philadelphia and New York, x + 
249 pp., 45 figs. 
Talbot, F. H. and M. J. Penrith 

1962. Tunnies and marUns of South Africa. Nature, vol. 193, no. 4815, pp. 558-559. 



PROCEEDINGS 

OF THE 

CALIFORNIA ACADEMY OF SCIENCES 

FOURTH SERIES 



Festschrift for George Sprague Myers 



Vol. XXXVIII, No. 8, pp. 139-156; 6 figs.; 3 tables December 31, 1970 



A NEW SPECIES OF GLANDULOCAUDINE 

CHARACID FISH, HYSTERONOTUS 

MYERSI, FROM PERU 

By 

Stanley H. Weitzman 
Smithsonian Institutio)i, Washington, D.C. 

and 

Jamie E. Thomerson 
Southern Illinois University, Eclzvardsville, Illinois 

INTRODUCTION 

On August 23, 1964, one of the authors (Thomerson), Jerry Anderson, 
Albert J. Klee, Emanuel Ledecky-Janachek, Winfield Rayburn, and Dr. Rich- 
ard L. Stone made a collection of fishes taken from a small stream tributary to 
the Pachitea River (Amazon drainage) at the northeastern outskirts of 
Tournavista, Province of Huanuco, Peru. Some of these were kept alive for 
experimental purposes and some were preserved. Among the fishes taken were 
representatives of the new species described here. 

Hysteronotus is a small genus of glandulocaudine characids most recently 
reviewed by Bohlke (1958) who described a new species, Hysteronotus hes- 
perus, amplified our knowledge of the only other known species, Hysteronotus 
megalostomus Eigenmann (1911), and redefined the genus. The characters of 
the new species described here and an analysis of additional specimens of H . 
megalostomus require a reevaluation of Bohlke's contribution. 

[139] 



140 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

Hysteronotus myersi Weitzman and Thomerson, new species. 

(Figures 1, 2, 3, 4, and 5.) 

Material. Holotype, a male USNM 203697, standard length 49.00 mm. 
(no. 13 in table 1) from a small stream directly tributary to Pachitea River 
(itself tributary to Ucayali River) at northeastern outskirts of Tournavista, 
Huanuco Province, Peru. Elevation approximately 200 meters. Paratj^es, 
originally in two lots, one lot of 8 specimens (nos. 1-4, 7-9, and no. 14 in table 
1) with same data as holotype. Second lot of 5 specimens (nos. 5, 6, and 
10-12 in table 1) raised in aquaria by Thomerson and bred from specimens 
in lot 1 and the holotype. Disposition of these lots is as follows: specimens 
nos. 5, 6, 7, 11, and 14 to Academy of Natural Sciences, Philadelphia (ANSP 
no. 112326 for nos. 5, 6, and 11, and ANSP no. 112325 for nos. 7 and 14); 
nos. 1, 2, 3, 4, 8, and 9 to United States National Museum, (USNM no. 
203698); nos. 10 and 12 to Tulane University Collections (TU no. 56456). 

Description. Proportions as thousandths of standard length appear in 
table 1. Body elongate, laterally compressed, especially in males; body depth 
just anterior to dorsal and anal fin 2.7-3.4 times in standard length. Predorsal 
body profile slightly convex with slight concavity at nape; concavity deepest 
at posterior termination of supraoccipital spine. Along base of dorsal fin, body 
surface slightly arched dorsally to accommodate inclinator and other muscles of 
fin. Posterior to dorsal fin, body profile nearly straight with gentle downward 
slope to adipose fin. Posterior to adipose fin, body profile a straight level line 
to procurrent caudal rays in males and a slightly downward slope to these 
rays in females (compare figs. 1-4). Ventral profile to anus usually gently 
rounded with steepest inclination ventral to jaws. Ventral profile protrudes 
ventrally its greatest distance at point ventral to midlength of adpressed pec- 
torals. At anal fin origin (anterior termination of fin base) body profile gently 
convex, more so in males, and slopes upward to beginning of caudal peduncle 
just posterior to posterior anal fin termination. At that point profile straight 
and level or sloping slightly downward to procurrent caudal fin rays. Caudal 
peduncle deeper in males, least depth in standard length 6.5-6.8 times in 
males and 7.5-8.5 times in females (compare figs. 1-4). 

Length of head 3.7-4.0 times in standard length, this proportion not chang- 
ing greatly in different sized specimens. Specimen 49.6 mm. (longest) and one 
28.3 mm. in standard length both with head 3.9 times in standard length. Eye 
rather large, somewhat larger in small specimens, 2.8-3.3 times in head length. 
Snout short, equal to, or shorter than, eye in specimens at hand, 3.3-3.9 times 
in head length. Snout appears proportionally longer in small specimens (table 
1). Least width bony interorbital 2.6-3.0 times in head length, always longer 
than snout length. 

Maxillary long, relatively slender, sloping ventrally and posteriorly to form 
an angle of 60-80 degrees to longitudinal axis of specimens. Maxillary length 



Vol. XXXVIII] WEITZMAN & THOMERSON : HYSTERONOTUS MYERSI 



141 



.a, 






S s 






I — oo ^ f^ *^ ^ 
oa 00 r— O CO 1^ 
ro I'i cv] rj- l« "Tf 



(V) r^l 00 "-I rf -1- 

r^ O O 00 --I t^ 

ro vO r-i -^ vO ■^ 

I I I I I I 

r^i "-I <~~a rf ^O rf 









o> 


O 


,-1 


o 


rO 


r-l 


■^ 




^o 


Tj- 


rf 


■o 


00 


VO 


t^ 


*"* 








rn 


Ul 


oq 


^ 


vo 




a 


















a 




q 














>. 




ON 


o 


r^i 


00 


<N1 


(^1 


^ 


^ 


^ 


-* 




00 


o 


tri 


00 


'"' 






ro 


ID 


Cvl 


^ 


lO 




"o 


















ffi 




vo 


















o 


rsi 


•* 


1— < 


t^ 


■^ 


^ 




«) 


ro 




o 


t^ 


f^ 





tr> CH- 



fo o 



(v^ in (V) ■^ ^O ^ 



in \o ^ '"' 



rO 


rO 


in 


<N1 


in 


ID 


Tf 
"* 


00 














t~0 










o 


in 



ro 1/^ rvi -^ lO tJ- 



00 ^ -"^ "^ "^ O 



(^ 




00 

m 


o 


in 


vn 








00 

in 




00 


in 


in 

rn 


in 

CO 


en 


O 

ON 

in 


oo 

ON 
CM 




O 

ON 

in 





O -^ O m 00 ON On 

t-^^ ro i^ t^ *^ t^ in 
re m CN ^ in ■^ 



ON On •'^ Ov) t^ »— < J^ 

CM o On t~- in r^ in 
CO in rg -rj- in Tf 



O -t^ 1^ O- NO 



00 --I NO ON On r^ ro 

o^i >— I 00 00 m ON NO 

ro in c-1 -^ in -^ 



CO rsl vO f^l On NO ^ 

r^i On r^ i^ in NO ro 

csi in r^i -^ in -^ 



r<> -^ OO CO •-< 
rO ON NO in O 
ro in r^ -^ NO 



x^inOONNO^— 'O^^OnO 
O ""j q J^ oo t~- in in in co 
in.-(rM^cot-~o6inro'— I 

NOro^f^i roONinoot^ON 



rn CO '— t r^ 

00 in in Tt 

I I I I 

00 t^ i-i rM 



I I I I 



O O O ^ 



O O O <-< 



in"— iT-Hr-v, -Hi— (NOf^NOro^^ 
x^ in ro cs] x^ i-o in I*^ X^— on ro 
T^t" .— I I— I t^i r-( <^i n o O O "— ' 



■^ ■^ CO NO 00 <V) O 

x^ T^ in r^ t^ ON ro 

cH CV) (N) O O O '"' 



<^!CiOO'-viONOOONT}-oO'n 
oo'^ioroinONOx^r^oo'^ 
■^T-i.-irNii-HrMrvlOOO'-* 



»-H rs] NO NO »— I ON 00 in "^ 
■^-HrvjNOinx^NOoOf^ 

■-Irsi.-H.-ll^lOOO'-l 



CO^oCC'^O''-)'— irO"— ION"* 

inco'^-csjcot^NOcox^oo^Nj 
T^i-H.-Hoii-i.-irsiOOO'-i 



OO'^I'-'OnOnOOOnOOnoO 
NO r^l ^o (^1 rvj On -^ CO NO CO (^ 
■^•-<^rs)rt--irv4 000'-' 



NOCOlnNONOOOONCO 

NOco^cNjcoONinoo 



O O NO rsj ON CN! X^ 
~~; O O 00 1^ On oq 
•-H (VI rsi O O O "-I 



oaNOONO'^'^NO'— I 
[ — c^Tj-r^rsioOinON 



NO00C00000'^00t-~^'*Tt- 

r^r^cocgcnooinoOX^ONrg 



r-'-ioO'^ooOooONOr^-* 
NOrocorvjcoOinONNOONCNi 



r<i O O O ■-< 



cor^NO-^oONOr^ONOOONi-H 
vO"— I'^'HrNiONNOoONOoOCNi 
Tj-i-H'-Hrg'-i.-Hr-jOOO'-i 



<-i M CM O O O >-i 



s 

t3 -S S" u > rt - 



C-. JO 



C 3 

3 Xi 



O '5 rt -Q O) _ 



Jj T3 



O o O o 



O 
O -O 



a 



a a 



Oj O O J3 



'„ ^ o o 



JZ JZ JS 



3 3 3 3 






aj O O O O 



C/3 O C/2 CA) CAl 



c)5 w' < 



-^ tl. tt t£ "fef -t C oju 

Q. c C C ■-; C cc C 



C C 

O a> 



142 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 



.„f«^r:^ftJ 






Nb 
^^^«^ 






Figure 1. Hysteronotus myersi, new species, holotype, USNM 203697, adult male, 49.0 
mm. in standard length. Small stream (tributary to Pachitea River, tributary to Ucayali 
River) at northeastern outskirts of Tournavista, Huanuco Province, Peru. 



(measured from tip of snout to posteroventral end of maxillary) 1.9-2.2 times 
in head length. Teeth 7-10, tricuspid, in single row on maxillary. Four speci- 
mens w^ith 7, three with 8, three with 9, and two with 10 teeth on one side. 
Teeth cover about 60-90 percent of free edge of maxillary. Premaxillary teeth 
in two series; outer row with 3 teeth except two specimens with 4 teeth on 
one side and 3 teeth on other side. Inner row with 4 tricuspid or quincuspid 
teeth in six specimens and 5 teeth in eight specimens. Usually 4 large, most 
often tricuspid, anterior teeth on each dentary (3 teeth on one side of one 
specimen). In large male specimens third tooth from anteromedian tooth larg- 
est and with secondary cusps reduced or absent. Sometimes other large den- 
tary teeth with reduced cu.sps. Large teeth followed by 9-13 abruptly smaller 
and usually tricuspid teeth. No teeth on vomer, palatines, or pterygoids. 

Fontanels almost absent, that part anterior to epiphyseal bar (often called 
frontal fontanel) not detectable, that part posterior to bar (often called pari- 
etal fontanel but almost always surrounded by frontal as well as parietal bones 
and supraoccipital) narrow, almost completely closed joint in all specimens. 
Gill rakers moderately short, pointed, longest less than /2 length of pupil, 
6-8 in upper limb, 10-12 on lower limb. Two specimens with total of 16, 
four with 17, four with 18, three with 19, and one with 20 rakers on entire 
first arch of one side. Circumorbital bones well ossified, covering entire cheek 
area, so-called "great suborbital" (actualh^ infraorbital 3) completely covers 
cheek, leaving no space between it and preopercle. Suprapreopercular process 
extends dorsally to level of dorsal fin of fourth infraorbital bone (postorbital of 
some authors). In large specimens posterior border of fourth infraorbital con- 



Vol. XXXVIII] WEITZMAN & THOMERSON : HYSTERONOTUS MYERSI 143 



,^-»SSiT'-"^ 



.-.Z 








Figure 2. Hysteronotus myersi, new species, paratype, USNM 203698, adult female 
32.6 mm. in standard length. Same data as holotype. 



tacts suprapreopercular process. Small individuals with space between these 
bones. Fifth infraorbital not in contact with preopercle. 

Scales of moderate size, cycloid \vith concentric circuli and about 8-15 
grooves or radii on the exposed posterior field. Lateral line complete, perfo- 
rating 39 scales in three specimens, 40 in four, 41 in three, 42 in four. Lateral 
line with slight ventral curve on side of body anterior to position of dorsal 
fin. Lateral line continues to caudal base along midline. Transverse scale 
rows between anterior bases of dorsal and anal fins 14-15, often 7 above and 
7 below lateral Une. Predorsal scale count 21-23; axillary scales present above 
pectoral and pelvic fins. Basal scale sheath at base of anal fin of about 27-29 
scales, usually 2 obvious horizontal rows anteriorly with some accessory scales. 
One longitudinal scale row along posterior third of anal fin base, and IV2 
rows at midregion of fin. Between bases of pelvic fin and anus, scales of both 
sides of body meet at midline in elongate median acute angle. Scales overlap 
acute midline angle only anteriorly near base of pelvic fins. No sharp keel 
between pelvic bases and anus. Area from anterior and posterior medial bases 
of pelvic fins along midventral line to isthmus, covered by scales. Ventro- 
lateral bases of pectoral fin without greatly enlarged scales. Figure 3 diagrams 
scales around caudal gland at base of lower caudal fin lobe. Two lateral line 
scales illustrated just dorsal to posterior base of gland. Glandular tissue and 
fossa-Hke structure of gland entirely supported by modified scales, fibrous 
connective tissue, and skin. 

Dorsal fin with ii, 9 rays in ten specimens, ii, 10 in four specimens; origin 
usually vertically over anterior base of anal fin, sometimes somewhat posterior 
to anterior anal fin base, nearer margin of opercle than base of caudal fin. 
Distance from tip of snout to anterior base of dorsal fin 1.7 1.8 times in 



144 CALIFORXIA ACADEMY OF SCIENCES IProc. 4th Ser. 




Figure 3. Hysteronotiis myersi, new species, holotype USNM 203697. 

Standard length. Dorsal fin profile rounded, not "straight topped" as reported 
for Hysteronotus hesperus by Bohlke (1958). Length of longest fin ray 
(= height of dorsal in table 1) 4.1-6.0 times in standard length; large males 
with greatest dorsal fin height (4.1 and 4.2 vs. 4.9-6.0 for all other specimens) 
(see also table 1). Height of dorsal fin appears sexually dimorphic, but rela- 
tively short in females and small males. 

Anal fin with v, 34 rays in two specimens, v, 35 in eight specimens, and 
V, 36 rays in four specimens. First unbranched ray not visible externally. 
Origin at or slightly behind midpoint of standard length. Distance from tip 
of snout to anal fin origin 1.6-1.8 (1.7 in eleven of fourteen specimens) times 
in standard length. Ventral margin of anal fin nearly straight in females, con- 
vex in males (compare figs. 1 and 2). Males with small dorsally recurved 
hooks on fourth through eleventh or twelfth branched anal fin rays (see fig. 1). 

Pelvic fin rays i, 6 in all specimens, distal end always reaching anterior 
basal termination of anal fin. Length of pelvics sexually dimorphic, 5.8 times 
in standard length in largest males, 6.3 in smaller male and 7.4-8.1 in females. 
Two types of contact organs present, bony hooks and bony spinelets. Hooks 
of one large, thick, hooklike excresence per ray segment. Spinelets of small, 
slender spicules of bone, one or more per ray segment. Spinelets easily broken, 
hooks not easily broken. Hooks confined to anal fin. Retrorse bony spinelets 
on males very small, and primarily on the first and second branched ray, even 
in largest male; not nearly as well developed or common as on Hysteronotus 
hesperus. Two to 3 or 4 spinelets per bony segment of each fin ray. 

Caudal fin with 10/9 principal caudal rays (17 branched rays) in all 
specimens; fin deeply forked. Males with small antrorse spinelets on dorsal 
edge of caudal rays, especially of lower lobe. No caudal spur. 

Vertebral counts 38-39 including ural segment. Two specimens with 16 
precaudals and 22 caudals, remainder (except for one abnormal specimen for 
which there is no count) with 16 jDrecaudals and 23 caudals. 

Color in alcohol. Humeral spot present, large, diffuse, and centered 
above fourth through sixth scales of lateral line. Single narrow, black, straight 
line extends from dorsal border of humeral spot to center of caudal peduncle 



Vol. XXXVIII] WEITZMAN & THOMERSON: HYSTERONOTUS MYERSI 145 




Figure 4. Hysteronotus myersi, new species, paratype ANSP 112326, adult female 30.9 
mm. in standard length. Bred from specimens collected at the type locality. 

where in males line arches dorsally to end at junction of center of upper caudal 
peduncle muscle mass with upper lobe of caudal fin (fig. 1). Line may be 
more diffuse than shown in fig. 1, or may be pale in some females as in fig. 2. 
Caudal blotch present, weak, sometimes absent as in fig. 1 ; weakly present in 
fig. 2. In male 36.5 mm. in standard length caudal blotch moderately well de- 
veloped at center of union of caudal fin with caudal peduncle. Anterior border 
of blotch diffuse but with some dark pigment extending onto central caudal 
rays. Never as much pigment as in Hysteronotus hcsperus. Compare figs. 1 
through 4 with fig. 2, plate 3 in Bohlke (1958) for H. hesperus and fig. 4, 
plate 58 in Eigenmann (1927) for H. megalostomus. Most of body of Hy- 
steronotus myersi pale brovm, slightly darker dorsally and lighter ventrally. 
Top of head dark brown with a narrow band of dark pigment extending from 
head to dorsal fin base. 

Color in life. One of us (Thomerson) has kept two pair of H. myersi 
in aquaria for several months. Their color may be summarized as follows. 
Females silvery with no prominent markings. JNIales with humeral spot and 
dusky stripe or band extending length of body. Both sexes with a distinct 
greenish iridescence. When males excited, lateral band darkens and 2 distinct 
pinkish spots appear at upper and lower base of caudal fin. 

Further aquarium notes. Fertilization is internal. Eggs slightly oval, 
approximately 1 mm. in diameter, and translucent. Eggs distributed on aquar- 
ium glass, plants, and rocks. More eggs appear attached to underside of plant 
leaves than on top. Very few eggs deposited near bottom of tank, usually in 
upper % of tank (5, 15, and 20 gallon aquaria). Spawning probably occurred 
in early morning and eggs appear deposited individually. 

Species name. This species is named in honor of George S. Myers in recog- 
nition of his long and continued interest in characid fishes, and his frequent 
and helpful council to students of this complicated but fascinating group. 

Type locality. Hysteronotus myersi is known only from the type locality, 
a small stream directly tributary to the Pachitea River (Amazon drainage) 
at the northeastern outskirts of Tournavista, Huanuco Province, Peru. Most 



146 



CALIFORNIA ACADEMY OF SCIENCES 



IProc. 4th Ser. 



Table 2. Measurements of Hysteronotus hesperus in thousandths of standard length. All 
specimens from eastern Ecuador. See Bohlke {19SS, p. 35) for localities; compare original 
numbers. 





1 


2 


3 


4 


5 


6 


7 


8 


Range 


Mean 














Holotyp( 


2 






Sex 


9 


9 


$ 


9 


S 


$ 


$ 


$ 






Standard length (mm.) 61.0 


62.3 


63.4 


68.2 


75.0 


76.0 


76.5 


81.8 






Greatest depth 


328 


320 


331 


320 


316 


349 


318 


352 


316-349 


326.75 


Snout to dorsal 


640 


644 


640 


650 


657 


663 


612 


621 


612-663 


640.88 


Snout to pectoral 


271 


265 


268 


258 


259 


286 


274 


271 


258-286 


269.00 


Snout to pelvic 


492 


475 


470 


469 


466 


470 


464 


454 


454-492 


470.00 


Snout to anal 


623 


623 


613 


640 


597 


627 


636 


606 


597-640 


620.63 


Eye to dorsal 


528 


518 


503 


532 


539 


542 


510 


514 


503-542 


523.88 


Anterior dorsal base 






















to caudal peduncle 


386 


401 


42 7 


384 


376 


423 


402 


412 


384-423 


401.37 


Depth of peduncle 


107 


106 


121 


113 


127 


126 


118 


127 


106-127 


118.13 


Length of peduncle 


134 


127 


114 


114 


131 


126 


131 


129 


114-134 


125.75 


Length of pectoral 


279 


279 


265 


264 


268 


271 


258 


— 


258-279 


269.14 


Length of pelvic 


139 


135 


142 


137 


136 


150 


140 


142 


135-150 


140.13 


Height of dorsal 


134 


156 


164 


180 


189 


191 


191 


183 


134-191 


173.50 


Length of head 


246 


229 


248 


232 


240 


268 


240 


252 


229-268 


244.37 


Diameter of eye 


069 


069 


060 


066 


065 


072 


059 


061 


059-072 


065.13 


Length of snout 


067 


075 


077 


070 


076 


083 


081 


073 


067-083 


075.25 


Bony interorbit 


090 


088 


087 


088 


084 


099 


089 


089 


084-099 


089.25 


Length of upper jaw 


107 


096 


096 


101 


105 


116 


094 


110 


094-116 


103.13 


Original number 


P304Piiooi : 


Pi607 


P308 


P309 Pil002 


P306 : 


P307 







of the specimens were taken from pools in an area of alternating shallow pools 
and riffles where the width of the stream varied from 1 to 5 meters and from 
a few centimeters to 0.5 meters deep. The bottom was gravel and sand, with 
a few snags and broken Umbs but no macrophytic aquatic plants. The stream 
was in a shallow ravine and was shaded by a dense canopy of small trees, 
brush, and vines. Downstream were several small waterfalls leading to an 
area of deeper boulder filled pools. Elevation at the tyi3e locality is approxi- 
mately 200 meters. A popular account of this locality is given by Klee 
(1965a). 

Fishes were not abundant, either above or below the waterfalls. Most of 
the specimens of Hysteronotus mycrsi were taken from midwater in the shal- 
low pools. Representatives of Rivulus peruanus Regan and loricariid catfishes 
were taken from the same pools, but the most abundant macroorganism was 
a river shrimp, Macrobrachium brazilensc (Heller). Klee (1965b) charac- 
terized the water at the type locaUty as ". . . . clear, clean, cool, moving 
water containing little or no vegetation. It is very soft, well oxygenated, 
and contains httle in the way of dissolved materials." These observations and 
the collection of the specimens of Hysteronotus myersi were made during the 



Vol. XXXVIIIJ WEITZMAX & THOMERSON: HYSTERONOTUS MYERSI 147 

last week of August and first week of September 1964, during the dry season. 
Relationships. Bohlke (1958) reviewed in detail our knowledge of Hy- 
stcronotus. At that time Bohlke distinguished the two known species, H. hes- 
perus and H. mcgalostomus, by contrasting 14 characters. In most of these 
characters, H. iiiycrsi appears closest to H. megalostoimis but differs from that 
species in many other respects. A new comparison is made of these 14 charac- 
ters plus additional characters based on new data for H. megalostomus, new 
counts and measurements of H. hcs penis (so that all counts and measurements 
are consistent), and data from H. viyersi. Tables 1, 2, and 3 present a com- 
parison of measurements as thousandths of standard length for the three species. 
Character 1, size of males: Standard length 63.4-81.8 mm. in Hysteronotus 
hesperus: 36.5-49.6 mm. in H. inycrsi: 29.0-41.8 mm. in //. megalostomus. 
All males at these various sizes appear fully adult. Both H. megalostomus and 
H. myersi appear to be relatively small species and the large adult males of 
H. myersi lived at least nine months in aquaria with little growth and are 
presumably large specimens of the species. Hysteronotus megalostomus may 
reach a larger size and perhaps these size differences between adult males of 
H. myersi and H. megalostomus do not reflect a real species difference. Char- 
acter 2, bony hooks on anal fin of male: H. hesperus with true hooks on last 
unbranched and first 8-9 branched rays. Bohlke (1958) reported hooks ex- 
tending back to third ray from posterior termination of fin; however, these 
are spinelets. Hysteronotus myersi with hooks confined to fourth through 
about twelfth branched rays, mostly on fifth through eleventh. Hysteronotus 
megalostomus with hooks on first through eleventh to twelfth branched rays. 
Character 3, pelvic fin rays: Rays i, 7 in H. hesperus; i, 6 in H. myersi and 
H. megalostomus. Character 4, humeral spot: Small, round, clearly defined in 
H. hesperus; diffuse and large in H. myersi: large, sharply defined, and 
vertically elongate in H. megalostomus. Character 5, outer and inner rows of 
premaxilliary teeth: Outer premaxillary teeth 4-6, usually 5 in H. hesperus: 3-4, 
usually 3 in H. myersi: and 3-5 in H. megalostomus. Inner premaxillary 
teeth 4-5, usually 4 in H. hesperus: 4-5, sUghtly more often 5, in H. myersi: 
and 5-6, usually 6 in H. megalostomus. Character 6, maxillary teeth: Teeth 
6-9 and very strong, dorsal teeth sometimes quincuspid, ventral teeth tri- 
cuspid in H. hesperus: 7-10 strong, tricuspid teeth in H. myersi: 5-6 strong 
(especially dorsally in large specimens) tricuspid teeth in H. megalostomus. 
Character 7, caudal fin of males split to its base: Not spUt to base in male 
of H. hesperus and H. myersi but split to base in H. megalostomus. Character 
8, pectoral rays: Normally i, 11 in H. hesperus: i, 9 in H. myersi; and i, 9 
(12 specimens) or i, 10 (6 specimens) in H. megalostomus. Character 9, lower 
limb gill rakers: 12 or, usually, 13 in H. hesperus: 10-12, usually 11, in H. 
myersi: and 10-12, usually either 11 or 12, in H. megalostomus. Character 10, 
eve in head length: 3.4-4.3 times in H. hesperus: 2.8-3.3 in H. myersi: and 



148 



C ALIFORM A ACADEMY OF SCIEXCES 



IProc. 4th Ser. 



Table 3. Measurements of Hysteronotus megalostomus in thousandths of standard length. 
Specimens 1-15 are from 3 to 4 km. northwest of Logoa Santa, Minos Gerais, Brazil. 
Specimens 16-18 are from a tributary of Rio das Velhas near Lagoa Santa, Minas Gerais, 
Brazil. 





1 


2 


3 


4 


5 


6 


7 


8 


9 


10 


Sex 


$ 


$ 


$ 


? 


$ 


? 


9 


$ 


? 


$ 


Standard length (mm.) 


28.4 


29.0 


29.3 


29.7 


29.8 


29.9 


29.9 


30.5 


31.3 


31.4 


Greatest depth 


274 


314 


324 


320 


329 


2 78 


291 


348 


294 


344 


Snout to dorsal 


612 


600 


604 


616 


584 


586 


612 


604 


600 


596 


Snout to pectoral 


264 


268 


272 


262 


278 


251 


264 


256 


268 


268 


Snout to pelvic 


431 


428 


447 


438 


449 


438 


438 


443 


447 


440 


Snout to anal 


545 


507 


550 


515 


547 


522 


553 


535 


565 


522 


Eye to dorsal 


510 


473 


485 


482 


474 


458 


478 


482 


489 


468 


Anterior dorsal base to 






















caudal fin base 


382 


42 7 


406 


407 


413 


398 


415 


400 


400 


420 


Depth of peduncle 


104 


128 


123 


118 


121 


100 


106 


125 


105 


124 


Length of peduncle 


139 


155 


154 


135 


138 


137 


134 


148 


137 


146 


Length of pectoral 


243 


241 


259 


239 


252 


224 


240 


229 


236 


246 


Length of pelvic 


104 


126 


137 


128 


121 


100 


113 


131 


102 


131 


Height of dorsal 


182 


186 


188 


192 


185 


177 


174 


194 


179 


185 


Length of head 


246 


231 


246 


246 


248 


244 


256 


239 


239 


242 


Diameter of eye 


083 


076 


085 


084 


084 


080 


080 


082 


083 


086 


Length of snout 


076 


079 


082 


081 


084 


077 


080 


082 


083 


083 


Bony interorbit 


090 


097 


097 


094 


098 


087 


097 


098 


096 


096 


Length of upper jaw 


lis 


107 


123 


118 


128 


107 


110 


121 


118 


111 


Color of pelvics 






black 




part 
black 






part 
black 







Table 3. Continued. 





11 


12 


13 


14 


15 


16 


17 


18 


Range 


Mean 


Sex 


$ 


$ 


9 


$ 


s 


$ 


$ 


$ 






Standard length (mm.) 31.6 


32.1 


32.2 


32.8 


41.8 


32.3 


34.0 


35.0 






Greatest depth 


352 


339 


298 


342 


316 


356 


338 


312 


274-356 


320.50 


Snout to dorsal 


614 


592 


582 


598 


630 


597 


606 


600 


582-630 


601.83 


Snout to pectoral 


272 


268 


254 


262 


271 


276 


270 


263 


251-278 


265.94 


Snout to pelvic 


446 


442 


432 


452 


440 


434 


456 


414 


414-456 


439.72 


Snout to anal 


544 


534 


534 


534 


524 


545 


535 


503 


503-565 


535.61 


Eye to dorsal 


478 


472 


475 


479 


505 


483 


462 


446 


446-510 


477.72 


Anterior dorsal base 


to 




















caudal fin base 


422 


436 


422 


424 


411 


440 


400 


374 


374-440 


410.94 


Depth of peduncle 


133 


122 


102 


125 


117 


124 


124 


111 


100-133 


117.33 


Length of peduncle 


149 


157 


137 


137 


146 


149 


144 


140 


134-157 


143.44 


Length of pectoral 


250 


— 


229 


2i2 


254 


251 


248 


223 


223-259 


241.41 


Length of pelvic 


111 


131 


096 


137 


127 


127 


118 


108 


096-137 


119.50 


Height of dorsal 


190 


181 


176 


183 


196 


179 


197 


177 


174-197 


184.50 


Length of head 


246 


242 


220 


250 


239 


242 


236 


228 


220-256 


241.11 


Diameter of eye 


089 


081 


078 


086 


084 


078 


082 


074 


074-089 


081.94 


Length of snout 


079 


075 


075 


079 


074 


074 


074 


069 


069-084 


078.11 


Bony interorbit 


101 


097 


090 


095 


083 


093 


088 


080 


080-101 


093.16 


Length of upper jaw 123 


115 


112 


116 


100 


118 


118 


114 


100-128 


115.22 


Color of pelvics 


black 


black 






> 


'ellow 


red 


black 







Vol. XXXVIII J WEITZMAX & THOMERSOX : HYSTEROXOTUS MYERSI 149 

2.8-3.1 in H. mcgalostomus. Character 11, length of upper jaw in head length: 
2.3-2.6 in H. hespcrus\ 1.9-2.2 in H. myersi; and 1.8-2.3 in H. megalostomus. 
Character 12, length of pelvics: 6.2-7.4 in males, 7.2-7.6 in females of H. hes- 
perus; 5.8-6.3 in males, 7.3-8.1 in females of H. myersi; and 7.3-9.2 in 
males, 7.8-10.4 in females of H. mcgalostomus. Character 13, anterior dentary 
teeth: Quincuspid in H. hesperus, tricuspid in H. myersi and H. megalostomus. 
Character 14, fine bony spinelets of male pelvic fins: Numerous and on both 
sides of ray segments, usually several per segment in H. hesperus: not numer- 
ous, 1 or 2 per segment (sometimes up to 4 in H. myersi) and on one side of 
ray only in both H. myersi and H. mcgalostomus. 

Other characters useful in comparing these species are as follows: Charac- 
ter 15, numbers of vertebrae: 40-42 vertebrae in H. hesperus \^^th 17 pre- 
caudals in all specimens, IZ caudals in one specimen, 24 caudals in two speci- 
mens, and 25 caudals in five specimens; 38-39 vertebrae in H. myersi, with 
16 precaudals in all specimens, 22 caudals in two specimens and 23 caudals 
in eleven specimens: 40-42 vertebrae in H. megalostomus with 15 precaudal 
vertebrae in almost all specimens and 25 caudal vertebrae in eight specimens, 
26 in seven specimens, and 27 in two specimens. One specimen of H. megalos- 
tomus with 14 precaudal and 27 caudal vertebrae. Character 16, scales around 
caudal fin: H. hesperus with 14 (15 in one specimen) longitudinal rows of 
scales around caudal peduncle, H. m.yersi and H. megalostomus vAih 18. 
Character 17, predorsal scales: This count difficult and inaccurate but H. 
hesperus with 23-25 scales, H. myersi \nth 20-23, and H. megalostomus with 
21-25. 

Character 18, tip of snout to dorsal fin origin in thousandths of standard 
length (see tables 1-3): Range of H. hesperus (612-663), mostly beyond 
ranges of other two species, (561-602) for H. myersi and (584-614) for H. 
megalostomus. Character 19, snout to anal distance in thousandths of standard 
length: Ranges of H. myersi (566-614) and H. megalostomus (507-565) 
partly contiguous, that of H. hesperus (597-640) begins at upper limit of 
range of H. myersi, not approaching that of H. megalostomus. Character 20, 
eye to dorsal distance in thousandths of standard length: Range of H. hesperus 
(503-542) nearly falls outside that of other two species (434-474 for H. myersi 
and 446-510 for H. megalostomus) . Character 21. distance between dorsal 
origin and base of caudal fin in thousandths of standard length: Ranges of 
H. hesperus (384-423) and H. megalostomus (374-440) broadly overlap; that 
of H. myersi (428-483) stands apart from that of H. hesperus and overlaps 
upper range of H. megalostomus. Character 22, length of caudal peduncle in 
thousandths of standard length: Ranges of H. myersi (131-151), and H. 
megalostomus (134-157) broadly overlap, while that of H. hesperus (114-134) 
barely overlaps their lower limit. Character 23, caudal gland: This gland is 
different in H. myersi and H. megalostomus (compare figs. 5 and 6). The 



150 C ALIFORM A ACADEMY OF SCIEXCES [Proc. 4th Ser. 

gland of H. he s perns is very similar to that of //. myersi (compare fig. 5 with 
fig. 6 in Bohlke 1958. Also see discussion below under Status of the Genus 
Hystcronotus). Character 24, caudal fin split to its base: The caudal fin is 
normally split to its base in Pseudocorynopoma doriae and Hysteronotus 
megalostomus but it is not spht in H. hesperus or H. myersi. 

The determination of the closest relative of H. myersi is difficult. As can 
be seen in the above characters, for example spinelets on the pelvic and caudal 
fins, number of ventral fin rays, number of outer row premaxillary teeth, 
number of cusps on maxillary teeth and large dentary teeth, number of pectoral 
rays, size of eye in relation to head length, length of the upper jaw, number 
of longitudinal rows of scales around caudal peduncle, proportional distance 
between snout tip and dorsal fin origin, proportional length of the caudal 
peduncle, and small scales at liectoral base, H. myersi more closely approaches 
H. megalostomus than it does H. hesperus. In a very few presumably im- 
portant characters, for example caudal fin not split to its base and caudal 
gland structure, H. myersi more closely approaches H. hesperus than it does 
H. megalostomus. 

In a few characters, for example length of pelvics in males, and relative 
distances between the dorsal origin and caudal fin base, H. hesperus and H. 
Diegalostomus are more similar to each other than either is to H. myersi. In 
some characters, for example in number of precaudal vertebrae, structure of 
glandular tissue within caudal gland, size and shape of the humeral spot, no 
bony hooks on first through third branched anal fin rays, relatively long 
pelvics in males, ventrally convex anal fin margin, and extremely rounded, 
convex male dorsal fin profile, H. myersi is unique and unlike either H. hes- 
perus or H . megalostomus. 

With our present unclear knowledge of the phyletic and genetic stability 
of the caudal fin organ, or gland, it is difficult to weigh the significance of 
this structure in showing a close relationship between H. hesperus and H. 
myersi in contrast to the many characters that indicate H. myersi is closer to 
H. megalostomus. The caudal glands of glandulocaudine characids are in 
need of detailed comparative study, both in their histology and gross struc- 
ture. The best review of this subject to date is by Nelson (1964). See Gery 
(1964, fig. 4) for figures of Glanduloeauda, and Nelson (1964, figs. 3-5) for 
figures of Pseudocorynopoma, Ar gyro pleura, Gephyrocharax, Landonia, Cory- 
nopoma, and Glanduloeauda. Eigenmann and flyers (1927. plates 84, 86, 
and 88) illustrated Corynopoma, Landonia, Pseudocorynopoma, and Gephyro- 
charax. Unfortunately, at present we do not know enough about either caudal 
glands or other characteristics of glandulocaudine characids to utilize these 
glands as valid, generic differences. The formation of the glandular tissue 
and scales in Hysteronotus megalostomus on the one hand, and H. hesperus 
and H. myersi on the other, is very different (compare figs. 5 and 6). The 



Vol. XXXVIII] WEITZMAX & THOMERSOX : HYSTERONOTUS MYERSI 151 




Figure 5. Lateral view of caudal gland of holotype of Hysteronotus inyersi. 



gland of the latter two species is surrounded in part by several modified scales 
and the glandular tissue lies over the lateral surface of two modified scales 
which curve dorsally over the glandular tissue forming a deep longitudinally 
oriented fossa. This fossa is open on its lateral and ventral surface. The gland 
of H. megalostomus is very different and has a very modified scale oriented 
ventrally around glandular material. The gland in H. megalostomus most 
closely resembles that of Pseudocorynopoma, see Eigenmann and Myers (1927, 
pi. 84, figs. 4-5). In gross dissection of H. megalostomus no obvious modified 
glandular tissue is present, but thickened skin lies over the dorsal surface of 
the ventral, furrowed scale, this skin being also attached medially to the fin 
rays. The same structure is found in Pseudocorynopoma doriae Perugia. The 
pouch of the gland in H. megalostomus extends anteriorly four to five scale 



152 



CALIFORNIA ACADEMY OF SCIENCES 



[Proc. 4th Ser. 




Figure 6. Lateral view of caudal gland of a specimen of Hysteronotus megalostomus 
34.0 mm. in standard length from a tributary to Rio das Velhas near La^oa Santa 
(19°39'S. longitude, 43°44'W. latitude), Minas Gcrais, Brazil. This is the male with red 
pelvics in table 3. 



rows between the scales just ventral to the lateral line and the musculature 
of the caudal peduncle. The external opening of the pouch is held lateral and 
open by the modified ventral scale and two elongate scales just anterodorsal to 
it. In H. myersi the pouch extends medially and anteriorly beneath five or 
six scale rows. The striated glandular tissue within the pouch turns sharply 
dorsally just within the pouch. This tissue ends under the area of the termina- 
tion of the lateral line. In H. hesperus the pouch and scale structure is about 
the same as in H. myersi] however, the apparent glandular tissue in the speci- 



Vol. XXXVIII] WEITZMAN & THOMERSON : HYSTEROXOTUS MYERSI 153 

mens at hand is not striated and only slightly thickened. It lies in the same 
area as the striated tissue in H. myersi. 

PoLYCHROMATiSM IN Hysteronotus megalostomus. Myers (1953) col- 
lected, preserved and labeled separately 3 males belonging; to H. megalostomus 
because one had black pelvic fins, one had red, and the other yellow pelvic fins. 
Bohlke (1958, pp. 39-42) suggested that these 3 fishes represented 3 closely 
related species. He devised a key to separate them using the characters dis- 
cussed below, but did not describe any of the three as a new species distinct 
from H. megalostomus. The senior author has reexamined these 3 specimens and 
compared them with 15 other specimens of H. megalostomus (table 3). 

Bohlke found the body depth different in the 3 males, 3.1 for the black- 
finned fish, 2.9 for the red and 2.8 for the yellow. Remeasurement of these 3 
specimens gives 3.1, 3.0, and 2.9 respectively, but relative differences are valid. 
In the additional collection of H. megalostomus investigated, the black-finned 
males have a depth of 2.8, 2.9, 3.0, 3.0, and 3.1. Two of these specimens 
(nos. 5 and 8 in table 3) have much less black on their pelvics than Bohlke's 
specimen (no. 18 in table 3). Three of the additional male specimens in table 
3 have colorless (color in life unknown) pelvics and a body depth of 3.0, 
3.1, and 3.2 (2.9 and 3.0 in the yellow and red specimens). These males in- 
clude specimens larger and smaller than Bohlke's. Thus body depth is not a 
function of body length in the sizes examined. Bohlke correlated the number 
of pectoral rays (i, 9 in red pelvics and i, 10 in yellow pelvics) with color 
and believed it may be a species difference. There is no information on the 
red and yellow color of the pelvics in the new collection; however, of all speci- 
mens available with black pelvics, three have 9 branched rays and three have 10 
branched rays; of the three colorless males, one has i, 10, two have i, 9. 
Bohlke found that the dorsal fin of the red-pelvic-finned male extended to 
the base of the adipose fin, but fell considerably short of the adipose fin in 
the yellow-finned male. In the additional males, the dorsal never reaches the 
adipose fin; however, the height of the dorsal fin varies considerably (table 
3). Specimen no. 10 has a dorsal fin proportionally almost as high as specimen 
no. 17, the red-finned male; the dorsal of no. 10 nearly reaches the adipose 
fin and dorsal fin length does not separate these fishes into two groups. Bohlke 
reported 3 maxillary teeth in the red-finned fish and 4 and 5 maxillary teeth 
in the yellow-finned fish. We confirm his counts but the number of maxillary 
teeth is variable in several of the specimens at hand. Two specimens had 1, 
five had 2, four had 3, and three had 4 maxillary teeth. One specimen is 
damaged. Bohlke counted 31 branched anal rays in the red-finned fish and 
53 rays in the yellow-finned fish. We again confirm Bohlke's counts, but in 
the additional specimens the branched ray counts vary from 29-33 and black- 
finned specimens exhibit this whole range of counts. Finally, Bohlke reported 
the yellow-finned fish with 43, and the red-finned specimen with 41 perforated 



154 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4tii Ser. 

lateral line scales. In the specimens at hand this scale count ranges from 40- 
45. 

All known male specimens of H. megalostomus have black pigment in the 
form of large melanophores on the body just dorsal to the pelvic fin rays; 
some have more of this than others. The amount of black pigment on the 
pelvic fins is variable and one specimen (not recorded as part black in table 3) 
has a few large melanophores on one pelvic fin. Perhaps the thickened fleshy 
interradial membranes unique to the pelvics of the yellow-finned fish studied 
by Bohlke are correlated with sexual activity and vary with sexual activity. 
In view of the above facts, we suggest that all the specimens examined, in- 
cluding those previously examined by Bohlke, belong to H. megalostomus, a 
single, somewhat variable, species with polychromatic pelvic fins in the males. 

STATUS OF THE GENUS HYSTERONOTUS 

Bohlke's definition of Hysteronotus (1958, pp. 33-34) includes the new 
species here described with little difficulty. In his key to the glandulocaudine 
genera (p. 44) Bohlke used one character of the genus, dorsal fin origin nearer 
the caudal base than to the eye, to separate Hysteronotus from four other rather 
unrelated genera. This statement is supported by data for H. megalostomus 
and H. he s per us but 12 out of 14 specimens of U. myersi (table 1) have the 
dorsal fin origin nearer the e3^e than the caudal base. We thus expand the 
definition of Hysteronotus to include fishes showing this character, however 
a revision of Bohlke's key to the glandulocaudine characid genera should be 
deferred until a complete and detailed review of the species involved is avail- 
able. 

A more questionable decision is that to include species with such diverse 
caudal gland structure (figs. 5 and 6) in a single genus. If caudal gland 
structure is strongly conservative in glandulocaudines then perhaps H. myersi 
and H. hcsperus should be generically separated from H. megalostomus, a 
species closer to Pseudocorynopoma in this character. On the other hand, the 
three species here referred to Hysteronotus share a number of unique characters 
and the differences in caudal gland structure may have little phyletic sig- 
nificance. We think it best to retain Hysteronotus as here defined with three 
known species, H. myersi, H. hesperus, and H. megalostomus, until a more 
penetrating analysis of the phyletic significance of characters found in glandu- 
locaudine characids can be made. 

ACKNOWLEDGMENTS 

We are indebted to Dr. James E. Bohlke of the Academy of Natural 
Sciences, Philadelphia, for his generous loans of specimens of Hysteronotus 
and for comments on the manuscript. Jerry Anderson, Albert J. Klee, Emanuel 
Ledecky-Janachek, Winfield Rayburn, and Dr. Richard L. Stone assisted 



Vol. XXXVIII] WEITZMAN & THOMERSON: HYSTERONOTUS MYERSI 155 

Thomerson in collecting the specimens here described. Dr. Alfred E. Smalley, 
Tulane University, kindly identified the river shrimp. The collection was made 
during Thomerson's tenure as a National Aeronautics and Space Administra- 
tion Predoctoral Fellow at Tulane University and collecting expenses were 
partially supported by a grant from the Society of Sigma Xi. The drawing 
for figure 2 was paid for by a grant from the Graduate School, Southern 
Illinois University. Figures 1 and 2 are by Marion Johnson; figures 5 and 6 
are by the senior author. Radiographs were prepared by Edgar N. GrambUn 
and Masaw L. Williams. 

LITERATURE CITED 

BoHLKE, James E. 

1958. Studies on fishes of the family Characidae, 14. A report on several extensive 
recent collections from Ecuador. Proceedings of the Academy of Natural 
Sciences of Philadelphia, no. 110, pp. 1-122. 

ElGENMANN, CaRL H. 

1911. New characins in the collection of the Carnegie Museum. Annals of the Car- 
negie Museum, no. 8, pp. 164-181. 
1927. The American Characidae. Memoirs of the Museum of Comparative Zoology 
at Harvard College, vol. 43, no. 4, pp. 311-428. 
ElGENMANN, Carl H. and George S. Myers 

1929. The American Characidae. Memoirs of the Museum of Comparative Zoology 
at Harvard College, vol. 43, no. S, pp. 429-574. 
Gery, Jacques 

1964. Glandulocauda terofoli sp. nov., un nouveau Poisson characoide de la Re- 
publique Argentine, avec une note sur la "glande" caudale des Stevardiidi. 
Opuscula Zoologica, Miinchen, no. 78, 12 pp. 
Klee, Albert J. 

196Sa. Peruvian Epic, Part V. The Aquarium Journal, San Francisco, vol. 36, no. 5, 

pp. 222-226, 231. 
1965b. Water analysis from the Peruvian Amazon. The Aquarium Journal, San 
Francisco, vol. 36, no. 9, pp. 420-426, pp. 432-435. 
Myers, George S. 

1953. Hints to fish importers. . . . No. 10 (A strange glandulocaudine characin from 
the Rio das Velhas). The Aquarium Journal, San Francisco, vol. 24, no. 6, 
page 137. 
Nelson, Keith 

1964. Behavior and morphology in the Glandulocaudine fishes (Ostariophysi, Characi- 
dae). University of California Publications in Zoology, vol. 75, no. 2, pp. 
59-152. 



PROCEEDINGS 

OF THE 

CALIFORNIA ACADEMY OF SCIENCES 

FOURTH SERIES 



Festschrift for George Sprague Myers 



Vol. XXXVIII, No. 9, pp. 157-162; 3 figs. December 31, 1970 



REDISCOVERY OF THE LORICARIID CATFISH, 

ACESTRIDIUM DISCUS HASEMAN, NEAR 

MANAUS, BRAZIL 

By 

Robert L. Hassur 
Division of Systematic Biology, Stanford University, California 94305 

More than fifty years ago John D. Haseman (1911, p. 319, pis. 50 and 51) 
described and figured a very small elongate loricariid catfish which he col- 
lected near Manaus, Brazil, as a new genus and species, Acestridium discus. 
He distinguished the genus from Farlowella principally on the basis of the 
expanded, disclike end of the snout and the presence of many series of delicate, 
spiny ridges (with intervening depressions) on all of the scutes. Haseman 
had 3 examples, the largest (holotype) was 72 mm. in total length. The t>^es 
are now in the Field Museum of Natural History, Chicago. With the excep- 
tion of a restatement by Miranda Ribeiro (1912) of Haseman's original de- 
scription and the inclusion by Gosline (1945) of the species in his catalog of 
Central and South American catfishes, the species, so far as is known, has not 
been reported again. 

During a recent visit. Dr. Jacques Gery presented the Stanford Collec- 
tion with 2 examples of A. discus (fig. 1 a,b) collected on 23 October 1965, 
by E. Fittkau and himself in a small tributary of the Igarape Castanha, which 
meets the right (southwestern) bank of the Rio Negro at a point two hours 
by boat upstream from Manaus. The specimens (SU 64202) measure 49 
and 51 mm. in total length. A search of the Stanford Collection revealed 
9 other specimens of the same species collected by Dr. Carl Ternetz in 1924 

[157] 



158 



CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 





Figure 1. Acestridiuni discus, SU 64202, 49 and 51 mm total length; a, ventral; b, 
dorsal; c, lateral. Scale 1 cm. 



from the Igarape do Mai Joana, also near Manaus. These 9 specimens (SU 
64095) range in total length from 48 to 65 mm. The total known range of 
the siDecies then includes 3 tributaries of the Rio Negro near Manaus. 

The specimens fit Haseman's description and illustrations with but one 
exception. The length of the snout, as usually measured from the anterior 
of the orbit to its tip, is about 2.25 rather than 4 times in the distance 
from its tip to the anus. Apparently Haseman measured the snout length 
differently, as this same measurement made on his drawings is about 2.25. 
Perhaps he, like Regan (1904, p. 303), used the distance between the tip 
of the snout and the anterior border of the naked area containing the mouth. 
This snout measurement, as determined from Haseman's drawings, does yield 
about 4 times in the distance from snout tip to anus. 

Several characteristics in Haseman's description require discussion. The 
large retrose hooks on the dorsal and ventral surfaces of the expanded tip 
of the snout (fig. 2) are set in 4 rows of 3-4 hooks. One row Hes on either 
side near the margin of the disc, the other parallel to it and situated a short 
distance from the margin. The first pelvic spine of many of the Loricariidae, 
including Fatiowella, is covered with many small hooks. In Acestridium, the 
structure is comblike (fig. 3) with enlarged toothlike hooks in about 3 series 
confined to the median surface of the spine, the rest of the spine being naked. 
With the fins erected, the hooks point toward each other, and, with the at- 



Vol. XXXVIII] 



HASSUR: ACESTRIDIUM DISCUS 



159 




Figure 2. Pelvic fin. 38 X. 



tached fin membranes and supporting rays curving inward toward the midline, 
the whole structure appears like a small basket. This structure was common 
to all specimens of both sexes examined. Although sex was not easily deter- 
mined because of the small size of the fish and, in some specimens, a lack 
of any recognizable feature of the gonads, the structure does not appear to 
be a secondary sexual character. Regan (1904, p. 198) describes the sexual 
characteristics in males of this family as being generally confined to enlarged 
bristles on the sides and top of the head or on the pectoral fins. But he makes 
no mention of the pelvic fins. 

The body (fig. lb) is finely striped with brown above, pale below (fig. 
la) with the black spots on the lower lateral sides (fig. Ic) forming a dark 
border in the abdominal region and continuing completely across on the snout 
and the caudal peduncle. A broad, brown, lateral stripe (fig. Ic) extends 
posteriorly on either side of the head from the snout through the eye and 
tapers to a point above the posterior edge of the pelvic base. Each of the 
rather bulging eyes bears a smooth-margined operculum on the dorsal part of 
the iris. 

Acestridium discus differs from all known species of Farlowella not only in 
the sharply expanded snout-tip with its series of hooks and the definite spiny 



160 



CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 




Figure 3. Expanded disclike snout with rows of retrose hooks. 60 X. 



Vol. XXXVIII] HASSUR: ACESTRIDIUM DISCUS 161 

ridges of the scutes, which are fewer in number (25-27) than in any known 
species of Farlowella (33-34), but also in its generally smaller size; the 
rounded caudal fin; the absence of filamentous extensions on the main, ter- 
minal upper and lower caudal rays; the rounded (rather than acutely pointed) 
dorsal, anal, and pelvic fins; the straight vertical margins of the lateral scutes 
of the long caudal peduncle; five rather than six plates between the dorsal 
plate and the supraoccipital; and in having the ventral surface of the abdomen 
covered with two rather than three series of plates. 

LITERATURE CITED 

GosLiNE, William A. 

1945. Catalogo dos Xematognath os de Agua-Doce da America do Sul E. Central. 

Boletim do Museu Nacional do Rio de Janeiro, N. S., Zoologica, no. 33, 138 

pp. 
Haseman, John D. 

1911. Descriptions of some new species of fishes and miscellaneous notes on others 

obtained during the expedition of the Carnegie Museium to Central South 
America. Annals of the Carnegie Museum, vol. 7, no. 304, pp. 315-328. 
Miranda Rlbeiro, Alepio de. 

1912. Loricariidae, Callichthyidae, Doradidae e Tricomycteridae. Comissao de Linhas 

Telegraphicas Estrategicas de Matto-Grosso ao Amazonas. Annexo No. 5, 
Historia Natural, Zoologica, 31 pp. 1 pi. Rio de Janeiro. 
Regan, C. Tate. 

1903. A monograph of the fishes of the family Loricariidae. Transactions of the 
Zoological Society of London, vol. 17, pp. 191-351. 



PROCEEDINGS 

OF THE 

CALIFORNIA ACADEMY OF SCIENCES 

FOURTH SERIES 



Festschrift for George Sprague Myers 



Vol. XXXVIII, No. 10, pp. 163-206. December 31, 1970 



THE AMPHIBIANS AND REPTILES 

OF AFGHANISTAN, 

A CHECKLIST AND KEY TO THE 

HERPETOFAUNA 

By 

Alan E. Leviton and Steven C. Anderson 
California Academy of Sciences, Scui Francisco 94118 

In 1950 the Academy received a collection of reptiles from Afghanistan. 
Though small, the collection was of considerable interest, having come from a 
little-visited region of the Dasht-i-Margo desert. It was reported on by Le\aton 
in 1959. Subsequently, significant collections of Afghanistan reptiles have been 
made, notably by John Gasperetti in 1961 (Leviton and Anderson, 1961a and 
1963), Richard and Erica Clark in 1964 and 1968 (Clark, Clark, Anderson and 
Leviton, 1969, and Clark and Clark, in preparation [for 1968 collections]), and 
the William Street Expedition of the Field Museum of Natural History in 1965 
(Anderson and Leviton, 1969). These, together with new materials acquired by 
other museums, some of which we have seen, have formed the basis of the follow- 
ing preliminary attempt at a checklist and key to the herpetofauna of the entire 
country. 

The growth of knowledge about the Afghan herpetofauna may be measured 
by the fact that in 1959 Leviton included 67 nominal species in his checklist. 
One additional species, Calotes versicolor, an error of omission at that time, 
should have been included to bring the total to 68. In the accompanying key 
and checklist, 101 nominal species and subspecies are listed, an increase of 50 

[163] 



164 CALIFORNIA ACADEMY OF SCIENCES LProc. 4th Ser. 

percent in the number of species currently known from that country. Of the 33 
new additions, five have been described as new since 1960. We believe it safe 
to say that our knowledge of the Afghan herpetofauna is still very incomplete 
and we expect that many interesting animals remain to be discovered, especially 
in the mountains of the Hindu Kush. 

We must emphasize that problems exist. For example, we are not satisfied 
with our treatment of the species of Eryx, Cyrtodactylus, or those of the genus 
Coluber. However, rather than postpone publication of this work indefinitely 
until all problems are solved, an unlikely event in any case, we beg the indulgence 
of our colleagues and hope they will find this account useful as a point of de- 
parture for extending their own researches. 

W^e have included in both the key and the checklist several nominal species 
whose presence in Afghanistan, though reasonable to expect, has yet to be 
authenticated. In the key these species are indicated by having an asterisk (*) 
immediately following the name; in the checklist the asterisk precedes the name. 

ACKNOWLEDGMENTS 

We are deeply in debt to a number of people who have generously given us 
assistance in the development of this work. First and foremost, we must empha- 
size that none of this would have been possible were it not for the dedicated and 
tireless field efforts of Mr. John Gasperetti and Mr. and Mrs. Richard Clark, 
the three being Field Associates of the California Academy of Sciences, and Mr. 
and Mrs. William Street, representing the Field Museum of Natural History. 
Their collections form the backbone of our work. In this regard Mr. and Mrs. 
Richard Clark deserve special mention, for they are themselves actively doing 
research in herpetology, yet they have not objected to our utilizing their material 
to the fullest extent possible. In addition we have drawn heavily upon compara- 
tive materials to be found in many museums and are thus grateful to those 
museums and their respective curators of herpetology: Dr. Robert F. Inger and 
Mr. Hymen Marx, Field Museum of Natural History; Dr. Richard G. Zweifel, 
American Museum of Natural History; Dr. James A. Peters, United States 
National Museum; Dr. Ernest E. Williams, Museum of Comparative Zoology, 
Harvard University; Dr. Robert C. Stebbins, Museum of Vertebrate Zoology, 
University of California at Berkeley; Dr. Donald Tinkle, Museum of Zoology, 
University of Michigan; Miss Alice G. C. Grandison, British Museum [Natural 
History] ; Dr. Josef Eiselt, Naturhistorisches Museum, Wien; Dr. Jean Guibe, 
Museum d'Histoire Naturelle, Paris; Dr. Ilja Darevsky, Zoological Institute, 
Academy of Sciences, Leningrad; Dr. F. W. Braestrup, Universitetets Zoologiske 
Museum, K0benhaven. 

At the Academy, several members of the Department's staff have assisted 
in the study, too: Charlotte Dorsey, Robert Drewes, James Harvey, Marilyn 
Kramer, and Linette Sabre. 



Vol. XXXVIII] LEVITON & ANDERSON: AFGHANISTAN HERPETOFAUNA 165 

In 1969 one of us (SCA) received a grant from the American Philosophical 
Society to facilitate the examination of Southwest Asian specimens in the U. S. 
National Museum, the American Museum of Natural History, and the Field 
Museum of Natural History. This trip also permitted the testing and revision 
of keys and so upgraded this work. 
KEY TO THE AMPHIBIANS AND REPTILES OF AFGHANISTAN 

1. Body, and limbs (if present), covered with scales or a shell (reptiles) 2 

Body and limbs without scales or shell (amphibians) _ 4 

2. Shell present (turtles) Testudo horsfieldii 

Shell absent - 3 

3. Limbs absent, eyes without movable lids (snakes) 79 

Limbs present, or if absent, eyes with movable lids (lizards) 9 

4. Tail present in fully metamorphosed individuals; hind limbs approximately same 

length as forehmbs; larvae resembhng adults, possessing teeth in both jaws 

(salamanders) Batrachnperus mustersi 

Tail absent in fully metamorphosed individuals; hind limbs considerably longer 
than forelimbs; larvae unlike adults, never possessing true teeth until metamor- 
phosis (frogs and toads) 5 

5. Large, raised gland (paratoid) between shoulder and eye; no maxillary teeth 6 

Paratoid gland absent; maxillary teeth present 7 

6. Tarsal ridge well developed, smooth; tympanum, when distinct, about half or less 

than half diameter of eye Bufo viridis 

Tarsal ridge marked by small tubercles ; tympanum % diameter of eye .... Bujo andersonii 

7. Skin of back with thickened dorsolateral ridge on each side; frequently with light 

vertebral stripe; toes not fully webbed (terminal phalanx of 4th toe free of 

web) Rana ridibunda 

Back without dorsolateral ridges; usually no light vertebral stripe; toes fully 
webbed 8 

8. Tympanum as large as eye or larger; heel reaching anterior to eye; prominent 

light stripe on hinder surface of thigh; males with external vocal sacs; males 

without pectoral glandular areas (mammata) Rana cyanophlyctis 

Tympanum smaller than eye; heel not reaching anterior to eye; no prominent light 
stripe on hinder surface of thigh; males without external vocal sacs; males with 
prominent pectoral glandular areas (mammata) Rana sternosignata 

9. Limbs absent; eyehds movable; deep longitudinal fold on each side of 

body Ophisaurus apodus 

Limbs present 10 

10. Pupil of eye vertically elliptical; skin soft, with granules, rarely imbricate scales, 

no paired shields on top of head, which is covered by granules; neither suborbital 
nor frontosquamosal arch present on skull; clavicles broadened, forming loop 

at inner end; tongue smooth or covered by threadlike papillae (geckos) 11 

Pupil of eye round or slightly oval ; skin covered by scales, plates, or granules, 
not soft ; if head not covered by paired plates, then by juxtaposed scales or 
granules; either suborbital and/or frontosquamosal arch present on skull; 
clavicles not broadened at inner end, or if clavicles broadened, tongue covered 
by imbricate, scalelike papillae, or by oblique folds 28 

11. Eyelids movable; digits not dilated; procoelous vertebrae Eublepharis macularius 

Eyelids immovable (spectacle); digits dilated or not; amphicoelous vertebrae 12 



166 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

12. Digits with expanded subdigital lamellae forming pads, lamellae paired 

Hemidaclylns jlaviviridis 

Digits without expanded subdigital lamellae ; lamellae not divided 13 

13. Digits with well defined lateral, combHke, flexible fringe of pointed scales 14 

Digits without lateral fringe of pointed scales, though scales may be denticulate, 

forming serrate border 19 

14. Dorsal scales small, granular or imbricate, intermixed with larger rounded 

tubercles IS 

Dorsal scales large, uniform, cycloid, imbricate 17 

15. Back with small, irregular, dark spots and dots; dorsolateral dark stripe extending 

from behind eye to just beyond forelimb or about % distance between fore- and 

hind limbs; tail with dark blotches above Crossobamon eversmanni 

Back with dark crossbars or longitudinal stripes 16 

16. Back with dark crossbars; hind limb reaching beyond axilla; ventral scales 

smooth Crossobamon lumsdeni 

Back with longitudinal stripes (or dots arranged in regular longitudinal rows) ; 
hind limb reaching axilla ; ventral scales keeled ; dorsolateral stripe extending 
from behind eye to, or almost to, insertion of hind limb; Hght dorsal stripe 
extending length of posterior % of tail, bordered by serrated brown dorsolateral 
stripes -. Crossobamon maynardi 

17. Cycloid scales on back extending on to hinder part of head Teratoscincus scincus 

Cycloid scales on back not extending beyond shoulders 18 

18. Cycloid scales on back feebly imbricate; about 100 scales round middle of 

body Teratoscincus microlepis 

Cycloid scales on back strongly imbricate; about SO scales round middle of 
body . _ Teratoscincus bedriagai 

19. Dorsal scales uniform, small, juxtaposed; granular, without enlarged tubercles 

present (Afghanistan specimens only) Alsophylax pipiens 

Enlarged dorsal tubercles present among granular scales 20 

20. Chin shields (postmentals) absent - 21 

Chin shields present 2 2 

21. Tail tapering gradually, covered below by small scales only; subdigital lamellae 

with several small tubercles, or denticulate on distal margin „.. Bunopus tuberculatus 
Tail cylindrical, very slender, of almost uniform diameter from base to tip, with 
median series of enlarged subcaudal plates; subdigital lamellae smooth 
Agamura persica 

22. Rostral excluded from border of nostril Agamura femoralis 

Rostral forming anterior border of nostril - 23 

23. Tubercles usually present among granules of lower surface of thigh, in short row 

of 1-6, some often in contact with posterior row of large imbricate scales; males 

with continuous series of preanal and femoral pores __. 24 

No subfemoral tubercles; males with preanal pores only 26 

24. 37-40 abdominal scales across middle of belly (19-25 in distance across belly equal 

to length of snout) ; series of pores separated on midhne by 1-2 scales not con- 
taining pores; 35-46 pores in males (total of both sides) Cyrtodactylus species 

23-34 abdominal scales across middle of belly (10-16 in distance across belly equal 
to length of snout) ; no separation between right and left series of pores; 24-41 
pores in males (total of both sides) 25 

25. 24-29 strongly keeled, nonmucronate trihedral or subtrihedral tubercles in para- 



Vol. XXXVIII] LEVITON & ANDERSON: AFGHANISTAN HERPETOFAUNA 167 

vertebral row from occiput to level of vent; males with 28-41 (32-40 in Afghan 
specimens examined) preanal and femoral pores (total of both sides) 

Cyrtodactylus fedtschenkoi 

20-23 strongly keeled, distinctly mucronate trihedral tubercles in paravertebral 
row from occiput to level of vent; males with 23-31 (24-29 in Afghan specimens 
examined) preanal and femoral pores (total of both sides) Cyrtodactylus caspius 

26. Subcaudal scales 1 head-width behind vent small, not enlarged and platelike (in 

distal part of tail, more than 2 head-widths posterior to vent, median series 
becomes enlarged, but much narrower than width of tail) ; tubercles on dorsal 
surface of tail arranged around middle of each caudal segment, not in terminal 

scale row Cyrtodactylus russowii''' 

Subcaudal scales 1 head-width behind vent enlarged, platelike, in single median 
series covering nearly full width of tail ; tubercles on dorsal surface of tail form- 
ing terminal ring of each caudal segment 27 

27. 25-47 abdominal scales across middle of belly (more than 12 scales across belly 

in distance equal to length of snout) Cyrtodactylus watsoni 

15-23 abdominal scales across middle of belly (7-10 scales across belly in distance 
equal to length of snout) Cyrtodactylus scaber 

28. No large paired shields on top of head, which is covered by granules or small scales 

or tubercles 29 

Enlarged paired plates on top of head (some granules may be present, but enlarged 
shields predominate) 54 

29. Venter covered by imbricate scales, not granules; tongue broad and short, smooth 

or covered with villose papillae, not deeply forked; dorsum covered by imbricate 

scales or combination of imbricate scales and granules (agamids) 30 

Venter covered by small granules or juxtaposed quadrangular scales; tongue deeply 
divided, long and slender, smooth, retractile into sheath at base; dorsum covered 
with numerous small granules or juxtaposed scales (varanids) 53 

30. Well marked dorsal crest, at least on neck Calotes versicolor 

No dorsal crest _. 3 1 

31. Femoral pores present; tail strongly depressed through most of length, dorsal 

surface of tail with transverse rows of very large spinous tubercles rounded 

at base 32 

Femoral pores absent; tail depressed only at base, without transverse rows of very 
large spinous tubercles rounded at base, although rings of large spiny scales may 
be present, forming more or less distinct caudal segments 33 

32. Back without greatly enlarged pointed tubercles; caudal spines small, 20-24 in 

cross-series at base of tail Uromastyx hardwickii 

Back with transverse series of large pointed tubercles; caudal spines large, 8-10 in 
cross-series at base of tail Uromastyx asmussi 

33. Tympanum exposed, i.e. ear opening visible 34 

Tympanum concealed or absent, no visible external ear opening 45 

34. Caudal scales in obhque rows, not forming rings ; tympanum deeply sunk 35 

Caudal scales forming more or less distinct rings ; tympanum large, superficial 36 

35. Dorsal scales subequal in size and disposed in regular rows Agama agilis 

Dorsal scales unequal, large dorsal scales twice as large as smallest, irregularly 

arranged Agama ruderata megalonyx 

36. Enlarged dorsal scales smooth or faintly keeled 37 

Enlarged dorsal scales strongly keeled 39 




168 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

37. Caudal segments with 2 whorls of scales 2 head-widths posterior to vent 

Agama caucasica 

Caudal segments with 3 whorls of scales 2 head-widths posterior to vent, or seg- 
mentation of tail indistinct 38 

38. 19-24 scales around tail at level of Sth complete whorl; scales on flanks distinctly 

larger than ventral and dorsolateral scales Agama badakhshana 

25-35 scales around tail at level of Sth complete whorl; scales on flanks small, not 
larger than ventrals, grading into granular dorsolateral scales Agama himalayana 

39. Caudal scales small, 30 or more in whorl S rows posterior to vent 40 

Caudal scales large, usually less than 30 in whorl 5 rows posterior to vent 41 

40. Forelimb and tibial portion of hind hmb covered above with scattered, greatly 

enlarged spinous scales surrounded by much smaller or granular scales, large 
scales neither grouped in patches nor imbricate ; largest scales on dorsum twice 
as large as largest ventral scales, vertebral and paravertebral groups of scales 

heterogeneous Agama nuristanica 

Forelimb and tibial portion of hind hmb covered with regularly arranged, enlarged 
imbricate and strongly keeled scales ; largest scales on dorsum not larger than 

ventrals, vertebral and paravertebral groups of scales homogeneous 

Agama tuberculata 

41. Flanks lacking enlarged scales Agama nupta 

Flanks with numerous enlarged scales intermixed with smaller scales 42 

42. Scales of chest and throat strongly keeled and mucronate Agama erythrogastra 

Scales of chest and throat smooth or weakly keeled, not mucronate 43 

43. Each caudal segment 1 head-length posterior to vent, with 2 whorls of scales 

Agama caucasica 

Each caudal segment 1 head-length posterior to vent with 3 whorls of scales 44 

44. Scales on snout and forehead smooth or faintly keeled; large mid-dorsal scales in 

4 longitudinal rows, intermixed with smaller scales; greatly enlarged dorsolateral 
and flank scales in small, separate clusters forming longitudinal rows 

Agama lehmanni 

Scales on snout and forehead strongly keeled; large mid-dorsal scales in regular 
longitudinal rows not intermixed with smaller scales, though vertebral row of 
small scales may separate enlarged scale rows into 2 paravertebral groups; no 
greatly enlarged dorsolateral scales; enlarged flank scales forming large patch 
on mid-flank Agama agrorensis 

45. Large cutaneous fold at corner of mouth Phrynocephalus mystaceus 

No large cutaneous fold at corner of mouth 46 

46. Dorsal scales heterogeneous, small scales intermixed with strongly enlarged scales .— 47 
Dorsal scales subequal, homogeneous (but in P. reticulatus clusters or single scales 

may appear to be of different size than surrounding scales because they are 
swollen and tubercular and with upraised posterior margins; if striking differ- 
ence is observed, see 47.) 48 

47. Enlarged dorsal scales flat, not tubercular, posterior border not sharply upturned; 

sides of back of head and neck with long, flexible, spinous or fringehke scales; 
both sides of 4th toe with long, well developed fringes ; tail without dark cross- 
bars, tip black on ventral surface in adults, in very small juveniles not black but 

with single black spot on ventral surface of tail Phrynocephalus liiteoguttatus 

Some enlarged dorsal scales nail-like, large portion of scale raised free of back ; 
sides of back of head and neck without long spinous or fringelike scales; 1 side 



Vol. XXXVIII] LEVITON & ANDERSON: AFGHANISTAN HERPETOFAUNA 169 

of 4th toe with short fringe; tail with dark crossbars always present at least on 
ventral surface, tip not black Phrynocephalus scutellatus 

48. Flexible, fringelike scales prominent in temporal region _ 49 

No fringelike scales in temporal region SO 

49. No crossbars on tail, tip of tail black; large, prominent black spots on back and 

top of head, group of 4 especially conspicuous in scapular region 

Phrynocephalus euptilopus 

Tail with distinct black crossbars on ventral surface, tip black; no conspicuous 
large, black spots on dorsum Phrynocephalus interscapularis 

50. Nasal shields separated by 1-3 series of scales ..__ SI 

Nasal shields in contact or partly separated 52 

51. Scattered scales or clusters of scales on dorsum with upraised posterior margins, 

often swollen, tubercular; scales of upper surfaces of limbs and midline of back 

prominently keeled Phrynocephalus reticulatus boettgeri 

No upraised, swollen scales on dorsum ; scales of upper surfaces of limbs and back 
smooth to indistinctly keeled in young, scales of limbs distinctly keeled in 
adults Phrynocephalus maculatiis 

52. Distinct, dark-margined, light dorsolateral stripe from posterior angle of eye along 

body onto tail; single, very elongate suborbital scale 2-3 times as long as ad- 
jacent scales Phrynocephalus clarkorum 

No light stripe along side of body: 3 suborbital scales of about equal size 
Phrynocephalus ornatus 

53. Tail round in cross-section, or slightly compressed posteriorly, without double- 

toothed crest above; abdominal scales in 110-125 transverse series from collar 

fold to groin Varanus griseus 

Tail compressed, with low, double-toothed crest above; abdominal scales in 90-110 
transverse series from collar fold to groin Varanus bengalensis 

54. Abdominal scales similar to dorsals; no femoral or preanal pores (skinks) 55 

Abdominal scales subquadrangular or quadrangular, in 8-18 longitudinal rows 

across venter, very distinct from dorsal granules; femoral pores present (except 

in Eremias aporosceles) (lacertids) 65 

55. Body elongate; limbs present but greatly reduced, 3-4 fingers, 3 toes 56 

Body not serpentine; 4-5 fingers, 5 toes 57 

56. Fingers 3 _ _ _ Ophiomorus tridactylus 

Fingers 4 Ophiomorus brevipes 

57. Lower eyelid scaly; palatine bones separated on midline of palate 58 

Lower eyelid with transparent disc or lids not movable ; palatine bones meeting 

on midline of palate 60 

58. 21-23 scales around body; postnasal present; single broad vertebral scale row, 

much broader than adjacent rows Eumeces taeniolatus 

26-30 scales around body; no postnasal; 2 median rows of dorsal scales broader 
than those on flanks -— 59 

59. 1 azygous postmental _ Eumeces blythianus* 

2 azygous postmentals — . Eumeces schneideri 

60. Eyelids immovable (spectacle) 61 

Eyelids movable — . 63 

61. Ear hidden beneath scales Ablepharus gray anus* 

Ear opening small but distinct _ 62 



170 CALIFORNIA ACADEMY OF SCIEXCES [Proc. 4th Ser. 

62. Frontoparietal single — Ablephariis pannonicus 

Frontoparietal divided Ablepharus bivittatus lindbergi 

63. Supranasals absent; pterygoid bones in contact anteriorly, palatal notch not reach- 

ing to level of centers of eyes Scmcella himalayana 

Supranasals present; pterygoid bones separated anteriorly, palatal notch extending 
forward to level of centers of eyes 64 

64. Prefrontals separated; 16-22 lamellae beneath 4th toe; dorsal scales feebly 

tricarinate or smooth Mabuya aurata* 

Prefrontals in contact; 12-16 lamellae beneath 4th toe; dorsal scales with 2-3 
strong keels Mabuya dissimilis 

65. Eyelids immovable; eye covered by spectacle ._.. _.. Ophisops jerdoni 

Eyelids movable - 66 

66. Nostril between 2 nasals and 1st upper labial Acanthodactylus cantoris subspecies 

Nostril not in contact with 1st upper labial ..„ 67 

67. Femoral pores absent Eremias aporosceles 

Femoral pores present 68 

68. Ventral plates in straight longitudinal series; lower nasal resting on 1st upper labial 

only; occipital shield in contact with interparietal Eremias giittidata watsonana 

Ventral plates in oblique longitudinal series ; lower nasal resting on 2-3 upper 

labials; occipital shield usually absent 69 

69. Subocular bordering mouth 70 

Subocular not bordering mouth 75 

70. Lateral scales of 4th toe in complete row length of toe, forming distinct fringe 71 

Lateral scales of 4th toe not forming distinct fringe 72 

71. Broad dark dorsolateral stripe from nostril through eye, along body and side of 

tail on each side, 1-2 additional narrower dark stripes medial to these on each 
side, remainder of dorsal dark stripes interrupted and anastomosing to form 
reticulate pattern, evident even in very young specimens ; 4th toe with 2 com- 
plete rows of subdigital scales, i.e. total of 4 scales counted around toe (except 

that extra scale may be present at a joint) Eremias scripta 

Dorsal pattern consisting of 7 dark stripes, outer dorsolateral stripes broadest, 
persisting unbroken in both adults and juveniles; 4th toe with single row of 
subdigital scales, i.e. total of 3 scales counted around toe (except that extra 
scale may be present at a joint) Eremias lineolala 

72. Back with dark stripes broader than interspaces in young and adults; no light 

ocelli on flanks nor spots contained within dark stripes, nor any tendency for 

flank stripes to break up — 73 

Back with dark stripes, breaking up into spots in adults; dark dorsolateral stripe 
containing white spots, or lateral and/or dorsolateral stripes breaking up into 
dark-margined ocelli 74 

73. 4th toe with 2 complete rows of subdigital scales and complete row of sharply 

pointed lateral scales, i.e. total of 4 scales counted around penultimate phalanx; 
collar scales small, usually only single median collar scale distinctly larger than 

adjacent gular scales anterior to collar Eremias fasciata 

4th toe with single complete row of subdigital scales and complete row of lateral 
to ventrolateral scales, i.e. total of 3 scales counted around penultimate phalanx; 
usually several collar scales distinctly larger than adjacent gular scales anterior 
to collar Eremias regeli 

74. Adults usually with black dorsolateral stripe, more or less continuous for at least 



Vol. XXXVIII] LEVITON & ANDERSON: AFGHANISTAN HERPETOFAUNA 171 

major portion of its length, containing white spots, black stripe contrasting 
strongly with dorsal color pattern; juvenile with 4 dark stripes on dorsum be- 
tween dorsolateral white-spotted stripes, vertebral stripe being white (dark 
stripes breaking up into 4 more or less regular rows of dark spots with 

age) Eremias velox persica 

Adults with dark interrupted dorsolateral black stripe forming ocelli with white 
spots, this dorsolateral pattern not contrasting strongly with interrupted dark 
stripes and spots of dorsum ; juveniles with 3 dark stripes on dorsum between 
white-spotted dorsolateral stripes, vertebral stripe being black, bifurcated on 
neck (dark stripes breaking up into several irregular rows of dark spots with 
age) Eremias velox velox 

75. 4th toe with distinct fringe on both lateral and medial sides, formed by complete 

row of sharply pointed lateral scales and complete row of similar medial scales; 

ungual lamellae of fingers and toes with prominent, flat, lateral expansions 76 

4th toe without distinct fringe; ungual lamellae without prominent lateral expan- 
sion 77 

76. Scales of flanks distinctly larger than those of back ; largest series of scales on 

lower surface of tibia only slightly broader than adjacent scales _„ Eremias grammica 
Scales of flanks not larger than those of back; series of broad plates on lower 
surface of tibia, more than twice as broad as adjacent scales Eremias acutirostris 

77. 2 series of femoral pores narrowly separated, space between series not exceeding 

1/4 length of each Eremias aria 

2 series of femoral pores widely separated, space between series at least % length 
of each 78 

78. 4th toe with single row of subdigital scales; usually distinct tympanic shield; 4th 

supraocular usually distinct Eremias intermedia 

4th toe with 2 rows of subdigital scales, internal much larger; tympanic scale 
usually small or indistinct; 4th supraocular usually indistinct ._- Eremias nigrocellata 

79. Ventral scutes not enlarged, same size as adjacent scales; eyes small, covered by 

scales 80 

Ventral scutes transversely enlarged ; eyes well developed 81 

80. Scales in 22-24 rows around body _- Typhlops vermicularis 

Scales in 14 rows around body Leptotyphlops blanfordi 

81. Ventral scutes narrower than full width of body; dorsal scale rows in more than 

35 longitudinal rows at midbody (Genus Eryx^) 82 

Ventral scutes as broad as full width of body; dorsal scales in 37 or less 
longitudinal rows 86 

82. Longitudinal dorsal scale rows less than 43 Eryx elegans 

Longitudinal dorsal scale rows 43 or more 83 

83. Width of interorbital space considerably greater than distance from posterior edge 

of eye to corner of mouth; front and upper surface of snout slightly convex; 
2nd upper labial usually higher than 3rd; ventrals without spots, or with widely 

separated dark spots Eryx jacukts* 

Width of interorbital space equal, less than, or shghtly greater than distance from 
posterior edge of eye to corner of mouth; front and upper surface of snout not 



1 Six currently recognized species of Eryx have been reported by various authors from Southwest Asia. 
Of these five have been at one time or another suggested as occurring in Afghanistan. A revision of the genus 
Eryx is long overdue. In lieu of such a revision we have included the five nominal species in our key; 
however we doubt that either E. jaculus or E. miliaris occurs in Afghanistan. See checklist for further 
comments. 



172 CALIFORNIA ACADEMY OF SCIENCES IProc. 4th Ser. 

convex; 2nd upper labial may be lower or higher than 3rd; ventrals as a rule 
with dark, confluent spots 84 

84. Width of interorbital space considerably less than distance from posterior edge of 

eye to corner of mouth; eyes directed upward; scales on tail smooth or with 

scarcely detectable keels; 2nd upper labial usually lower than 3rd Eryx miliaris* 

Width of interorbital space equal, slightly less than, or slightly greater than distance 
from posterior edge of eye to corner of mouth ; eyes directed laterally ; scales on 
tail with prominent keels, at least in adults 85 

85. Scales on body smooth, those on tail and on sides near anal region keeled; end of 

tail much narrower than head; no distinct bands on body or tail, but dark 

blotches and irregular markings present Eryx tataricus 

Scales of body and tail more or less distinctly keeled; tail extremely blunt, often 
as wide as head; unicolored or with series of distinct dark bands on tail, some- 
times on body, especially evident posteriorly Eryx johnii 

86. Top of head covered by numerous small scales, none arranged to form regular 

large plates; enlarged fangs present 87 

Top of head covered by 8-9 large plates disposed in regular pattern ; enlarged fangs 

present or absent . __. 90 

87. Lateral scales in oblique series 88 

Lateral scales in straight longitudinal series 89 

88. Subcaudals single; keels of lateral scales serrated; ventrals without strong lateral 

keel Echis carinatus 

Subcaudals paired ; keels of scales not serrated ; ventrals with strong lateral 
keel Eristicophis macmahont 

89. Supraocular "horn" present, surrounded by small scales; supranasal sac present, 

opening into upper part of nostril Pseiidocerastes persicus* 

Supraocular "horn" absent; no supranasal sac Vipera lebetina 

90. Facial pit present between nostril and eye .— 91 

No facial pit present - 92 

91. Temporals, posterior upper labials fused Agkistrodon himalayanus 

Temporals, posterior upper labials not fused Agkistrodon halys 

92. Loreal absent, nasal shield in contact with preocular; dorsal scales in 19-25 

longitudinal rows .— Naja oxiana 

Loreal present, nasal shield not in contact with preocular; if loreal absent, dorsal 
scales in less than 17 longitudinal rows 93 

93. Series of subocular scales separating labials from eye border; scales in 2 7-33 longi- 

tudinal rows at midbody; 10-13 upper labials Spalerosophis diadema 

One or more upper labials bordering eye, or if subocular scales present excluding 
labials from eye border, then scales in less than 2i longitudinal rows — - 94 

94. Scales in 15 longitudinal rows at midbody 95 

Scales in 17 or more longitudinal rows at midbody 96 

95. Loreal usually absent; temporals 1-|-1; subcaudals more than 60; head, nape with 

black crossbars or entirely black above Eirenis persica* 

Loreal present; temporals l-|-2; subcaudals less than 60; brown crossbars on head 
chevron-shaped — - — -^ Oligodon iaeniolatus* 

96. Dorsal scales keeled (except outer row usually smooth), in 19 longitudinal rows 

at midbody; ventrals less than 180; 1-2 anterior temporals — 97 

Dorsal scales smooth, or if keeled, then ventrals more than 180, 2-3 anterior 
temporals 98 



Vol. XXXVIII] LEVITON & ANDERSON: AFGHANISTAN HERPETOFAUNA 173 

97. 1 anterior temporal - _ Natrix tessellata 

2 anterior temporals .- - -- -- - Xenochrophis piscator 

98. Scales in 17 longitudinal rows at midbody 99 

Scales in 19-25 longitudinal rows at midbody - - 103 

99. Pupil of eye vertically elliptical; black above with white or yellowish cross- 

bars Lycodon striatus bicolor* 

Pupil of eye round; color not as described above — 100 

100. Anal single; 1 anterior temporal; 8 upper labials Psammophis leithi* 

Anal divided; 2 anterior temporals; 9 upper labials 101 

101. No longitudinal stripes on body and tail; no longitudinal stripe through eye; 

maxillary teeth 20-38, posteriormost not grooved Ptyas mucosus 

Color pattern consisting of longitudinal stripes (sometimes indistinct) ; longitudinal 
dark stripe through eye; maxillary teeth 10-13, posteriormost enlarged, 
grooved 102 

102. Caudals 72-115; 3 labials entering eye (4th to 6th); well defined longitudinal 

markings on top of head Psammophis lineolatus 

Caudals 118-134; 2 labials entering eye (5th and 6th); markings on top of head 
broken up into smaller blotches and spots -- Psammophis schokari 

103. Scales in 19 longitudinal rows at midbody --- 104 

Scales in 21-25 longitudinal rows at midbody 108 

104. Pupil vertically elliptical, or if pupil ivide open and rounded, then rostral 

projecting 10^ 

Pupil round; rostral broadly rounded - 106 

105. Usually 1 prefrontal; anal single; at most 1 upper labial bordering eye 

Lytorhynchus ridgewayi 
2 prefrontals; anal divided; upper labials not bordering eye Lytorhynchus maynardi 

106. Subocular present; 1 upper labial bordering eye Coluber karelinii 

No suboculars; 2 upper labials or more bordering eye 107 

107. Scale rows 11-13 just anterior to vent; ventrals 205-244; subcaudals 124- 

135 Coluber rhodorhachis 

Scale rows 13-15 just anterior to vent; ventrals 199-211; subcaudals 82- 
119 Coluber ventromaculatus* 

108. Pupil vertically elliptical; anal single Boiga trigonata melanocephalus 

Pupil round ; anal divided 109 

109. 2 preoculars, 1 subocular; 21-23 scale rows at midbody; 9 (rarely 8 or 10) 

supralabials ; posterior maxillary teeth longest; frontal with crescentic lateral 

margins Coluber ravergieri 

1 preocular, 1 small subocular; 23-25 scale rows at midbody; 8 (rarely 9) 
supralabials; anterior maxillary teeth longest; frontal with straight lateral 
margins Elaphe dione 

CHECKLIST OF THE AMPHIBIANS AND REPTILES 
OF AFGHANISTAN 

In the following list, species are arranged alphabetically within genera, genera 
within families, etc., and no systematic relationships are implied by the arrange- 
ment. 



174 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

Class AMPHIBIA 
Order CAUDATA 
Family Hynobiidae 
Genus Batrachuperus Boulenger 

Batrachupems Boulenger, 1878, Bull. Soc. Zool. France, vol. 3, pp. 71-72 (type species: 
Salamandrella sinensis Sauvage, by monotypy). 

Batrachuperus mustersi Smith. 

Batrachyperus mustersi Smith, 1940, Ann. Mag. Nat. Hist., ser. 11, vol. 5, pp. 382-383 
(type locality: mountain streams of Paghman Range, above Paghman, Afghanistan, 
9000-10,000 feet elevation; holotype: British Museum no. 1940.3.1.1). 

Distribution. Known only from the type locality. 

Order SALIENTIA 

Family Bufonidae 

Genus Bufo Laurenti- 

Bufo Laurenti, 1768. Synops. Rept., p. 25 (type species: Bufo viridis Laurenti, 1768, by 
subsequent designation by Fitzinger, 1843). 

Bufo andersonii Boulenger. 

Bujo andersonii Boulenger, 1883, Ann. Mag. Nat. Hist., ser. S, vol. 12, p. 161 (type locality: 
Ajmere, Rajputana [restricted by Leviton, Myers, and Swan, 1956, Occ. Pap. Nat. Hist. 
Mus. Stanford Univ. no. 1, p. 4] ; 3 syntypes in British Museum). 

Distribution. All of northern India at low elevations, from the Ganges 
Basin through Rajputana, Punjab, and Sind, to southern Afghanistan, north to 
Kashmir and Nepal; southern and eastern Arabia. In Afghanistan it is known 
from south of the Hindu Kush in the southeastern part of the country and along 
the Helmand River, west to the Seistan Basin. To 4500 feet elevation at 
Kandahar and Khost. 

Bufo viridis Laurenti. 

Bufo viridis Laurenti, 1768, Synops. Rept., p. 27, pi. 1, fig. 1 (type locality: Vienna, 
Austria) . 

Distribution. From southern Sweden and eastern France over all of Europe 
(except Iberian Peninsula) eastward to Mongolia, south in Central Asia to 



2 Clarification of the type species of tfie nominal genus Bujo: 

Bujo vulgaris Laurenti is cited as type species of the nominal genus Bujo by Stejneger (1907) and sub- 
sequent authors. This is in error. Bujo vulgaris was described by Laurenti. There is no indication that it 
is proposed as a substitute name for Rana bujo Linnaeus to avoid tautonomy, though this is not unlikely. 
Nevertheless it is subjective, there being no objective basis for this assumption. Also, inasmuch as B. vulgaris 
is not accompanied by a synonymy in which the Linnaean name is cited, nor are the figure references cited by 
Laurenti the same as those cited by Linnaeus, it is not possible to claim that B. vulgaris is a junior objective 
synonym of R. bujo Linnaeus, there being no indication that the two are based on the same type. Therefore, 
Bujo vulgaris Laurenti is type species of the genus Bujo only by reason of the fact that Stejneger so designated 
it in 1907. However, Stejneger was preceded by Fitzinger who, in 1843, designated the type species of 
Laurenti's genus as Bujo viridis Laurenti. To the best of our knowledge this constitutes the first acceptable 
designation of the type species of this nominal genus. 



Vol. XXXVIin LEVITON & ANDERSON: AFGHANISTAN HERPETOFAUNA 175 

Tibet and the Himalayas, throughout Southwest Asia, parts of Arabia, and the 
northern Sahara as far west as Morocco. Throughout almost the whole of the 
Euro-Siberian, Irano-Turanian and Mediterranean regions. To 15,000 feet in 
the Himalayas. 

Family Ranidae 
Genus Rana Linnaeus 

Rana Linnaeus, 1758, Syst. Nat., ed. 10, vol. 1, p. 210 (type species: Rana temporaria 
Linnaeus, 1758, by subsequent designation by Fitzinger, 1843). 

Rana cyanophlyctis Schneider. 

Rana cyanophlyctis Schneider, 1799, Hist, amph., vol. 1, p. 137 (type locality; eastern 
India). 

Distribution. From Thailand to Nepal and Ceylon, north to Kashmir and 
the Himalayas, and west through southern Afghanistan and Baluchistan to 
eastern Iran. It is also recorded from southern Arabia. In Afghanistan it occurs 
south of the Hindu Kush to elevations of about 4500 feet. It has been taken 
west of Dilaram and is known from the Iranian portion of the Seistan Basin. 

Rana ridibunda ridibunda Pallas. 

Rana ridibunda Pall.as, 1771, Reise versch. Prov. russ. Reich, vol. 1, p. 458 [not seen] (type 
locality: Gurjew [Gurev], USSR, north coast of Caspian Sea [restricted by Mertens 
and MuLLER, 1928, Abh. Senckenberg. Naturf. Ges., vol. 41, p. 20]). 

Rana ridibunda ridibunda Mertens, 1925, Abh. Senckenberg. Naturf. Ges., no. 39, p. 55. 

Distribution. The whole of Europe to 60° N. except northwest and central 
Italy; western Asia as far east as northern West Pakistan, Afghanistan, and 
eastern Turkestan; North Africa and Arabia as far south as the Hejaz. In 
Afghanistan it is known from the region north of the Hindu Kush, penetrating 
at least as far south and east as Paghman. It has been recorded from Iranian 
Seistan, however. 

Rana sternosignata INIurray. 

Rana sternosignata Murray, 1885, Ann. Mag. Nat. Hist., ser. 5, vol. 16, pp. 120-121 (type 
locality: Quetta, West Pakistan). 

Distribution. The Quetta Plateau in Baluchistan, north at least to Kabul 
in Afghanistan, and Kashmir. Known locaUties in Afghanistan lie between 4500 
and 8000 feet elevation. 

Class REPTILIA 

Order CHELONIA 

Family Testudinidae 

Genus Testudo Linnaeus 

Testudo Linnaeus, 1758, Syst. Nat., ed. 10, vol. 1, p. 197 (type .species: Testudo graeca 
Linnaeus, 1758, by subsequent designation by Fitzinger, 1843, Syst. Tept., p. 29). 



176 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

Testudo horsfieldii Gray. 

Testudo horsfieldii Gray, 1844, Cat. tort. croc, amphisb. British Mus., p. 7 (type locality: 
Afghanistan; holotype in British Museum). 

Distribution. From the northeast shores of the Caspian Sea eastward 
across Kazakhstan to Lake Zaysan and thence southwestward to Afghanistan, 
Waziristan, Baluchistan, and eastern Iran. To at least 8000 feet elevation in 
Afghanistan. 

Order SQUAMATA 

Suborder Sauria 

Family Agamidae 

Genus Agama Daudin 

Agama Daudin, 1802, Hist. nat. Rept., vol. 3, p. 333 (type species: Lacerta agama Linnaeus, 
1758, by absolute tautonomy). 

Agama agilis Olivier. 

Agama agilis Olivier, 1807, Voy. Emp. Otho., vol. 4, p. 394, and atlas, pi. 29, fig. 2 [not seen] 
(type locality: neighborhood of Baghdad, Iraq; syntypes: Paris Museum no. 5708 [2]). 

Distribution. Western Punjab; West Pakistan; Afghanistan; Iran; Asian 
steppes of the USSR (coasts of the Caspian Sea, east to the Tarbagatai, and 
north to the steppes of the lower reaches of Irgiz) ; Iraq. To about 7500 feet 
elevation in Afghanistan. 

Agama agrorensis (Stoliczka). 

Stellio agrorensis Stoliczka, 1872, Proc. Asiatic Soc. Bengal, pp. 128-129 (type locality: 

Sussel Pass, at the entrance to the Agror Valley, Hazara District, NW. Punjab, West 

Pakistan; 9 syntypes in British Museum). 
Agama agrorensis, Boulenger, 1885, Cat. liz. British Mus., vol. 1, p. 363. 

Distribution. Punjab (Agror, or Oghi Valley) : Kashmir (Jhelum Valley, 
Chilas) ; Chitral (Arandu) ; Afghanistan (valley of the Kabul River). To at 
least 6000 feet elevation. 

Agama badakhshana Anderson and Leviton. 

Agama badakhshana Anderson and Leviton, Proc. California Acad. Sci., ser. 4, vol. 37, pp. 
32-35, figs. 6-7 (type locality: Mazar-i-Sharif, Afghanistan; holotype: Field Museum 
of Natural History no. 161108). 

Distribution. Known from three localities, all in Afghanistan: Mazar-i- 
Sharif, and 64 miles east of Faizabad, both on the northern side of the Hindu 
Kush, and Paghman, on the southern side. 

Agama caucasica (Eichwald). 

Stellio caucasicus Eichwald, 1831, Zool. spec. Ross. Polon., vol. 3, p. 187 (type locality: 

Tiflis and Baku, Caucasus, USSR). 
Agama caucasica, Boulenger, 1885, Cat. liz. British Mus., vol. 1, p. 367. 



Vol. XXXVIII] LEVITOX & ANDERSON: AFGHANISTAN HERPETOFAUNA 177 

Distribution. Southeastern Caucasus and northeastern Turkey, east 
through the northern and mountainous regions of Iran, Transcaspian provinces 
of the USSR to Chubek, Tajikistan in the east, Afghanistan, and Waziristan and 
Baluchistan, West Pakistan. To 9000 feet in Afghanistan. 

Agama eiythrogastra (Nikolsky). 

Stellio erythroi^a^iter Xikolsky, 1896, Ann. Mus. Zool. Acad. Imp. Sci. St. Petersbourg, vol. 

1, pp. 370-371 (type locality: Kalender .^bad and Ferimun eastern Iran; syntypes: 

Zoological Institute Leningrad nos. 8759, 8760). 
Agama erythrogastra, Nikolsky, 1915, Faun. Russie, vol. 1, pp. 119-121. 

Distribution. Northeastern Iran, in the vicinity of Mashhad, southeastern 
Turkmen, northern Afghanistan, south through the mountain passes to Paghman; 
3000-8000 feet elevation. 

Agama himalayana himalayana (Steindachner). 

Stellio himalayanus Steindachner, 1867, Reise osterr. Fregatte N^ovara., Zool. Teil., Rept., 
vol. 1, p. 22, pi. 1, fig. 8 (type locality: Leh and Kargil, Ladakh frontier district, 
Kashmir ; syntypes in Vienna Museum ) . 

Agama himalayana, Boulenger, 1885, Cat. liz. British Mus., vol. 1, p. 362. 

Distribution. Himalayas, Trans-Himalayas, southern Tibet, Hindu Kush, 
the ridge system of Pamiro-Alai west up to the Pamir inclusive, and southern 
part of Tien Shan; not known north of the Fergana Valle}^ Chitral, Kashmir, 
and Ladakh ; known from Afghanistan only in the northeast. 

Agama lehmanni (Nikolsky). 

Stellio lehmanni Nikolsky, 1896, Ann. Mus. Zool. Acad. Imp. Sci., St. Petersbourg, vol. 1, 
p. xiv (type locality: Fergana and Bokhara, USSR; syntypes in Zoological Institute, 
Academy of Sciences, Leningrad). 

Agama lehmanni, Bedriaga, 1907, Wissenschaft. Result. N. M. Przewalski Central Asien 
Reisen, vol. 3, Amphib. u. Rept., p. 126, pi. 2, fig. 2. 

Distribution. The mountains of the southeastern part of Central Asia north 
to the Fergana Valley, west to the Nura Tau and Kugitang ridges, east to the 
Darvaz Ridge, south to northern Afghanistan (Terentjev and Chernov, 1949, p. 
148). Alazar-i-Sharif, at about 1500 feet elevation, appears to be the only 
documented record for Afghanistan. In the USSR it is found up to 11,000 feet 
elevation. 

Agama nupta De Filippi. 

Agama nupta De Filippi, 1843, Giorn. Inst. Lomb. e Bib. Ital., vol. 6, p. 407 (type locality: 
Persepolis, Iran; holotype in Milan) |not seen!. 

Distribution. West Pakistan; Afghanistan; Iran; Iraq. It appears to be 
confined for the most part to the southern margins of the Iranian Plateau, 
ranging from 1000-8000 feet elevation in Iran. In Afghanistan it is known only 
from the areas south of the Hindu Kush, up to 5000 feet elevation. 



178 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4tii Ser. 

Agama nuristanica Anderson and Leviton. 

Agama nuristanica Anderson and Leviton, Proc. California Acad. Sci., ser. 4, vol. 37, pp. 

39-42, fig. 8 (type locality: Kamdesh, Xuristan, Afghanistan; holotype: Field Museum 

of Natural History, no. 161136). 

Distribution. Known only from the type locality in Nuristan, eastern 
Afghanistan, on the southern side of the Hindu Kush at 1342 meters elevation. 
Smith's (1935, pp. 214-216) record of .4. tuberculata for Kabul may refer to 
this species. 

Agama ruderata Olivier. 

Agama ruderata Olivier, 1807, Voy. Emp. Otho., vol. 2, p. 429, pi. 29, fig. 3 (syntypes from 
Persia and northern Arabia; syntype: Paris Museum no. 2610) [not seen]. 

Trapelus megalonyx Gunther, 1864, Rept. British India, p. 159, pi. 14, fig. C (type locality: 
probably Afghanistan; holotype: 9 in British Museum). 

Distribution. North Arabian Desert (Syria, Jordan, Turkey, Iraq, but 
probably not Saudi Arabia), Iran, southern Afghanistan, and northv^^estern West 
Pakistan. Northward it reaches the southern shore of the Caspian Sea and the 
southeastern Transcaucasian region of the USSR; to about 7500 feet elevation 
in Afghanistan. 

Elsewhere (Clark, Clark, Anderson, and Leviton, 1969, pp. 292-294) we 
have commented upon the problems surrounding the proper allocation of the 
Afghan specimens. For the present, both A. megalonyx and A. r. baluchiana are 
considered synonyms of A. ruderata. 

Agama tuberculata Gray. 

Agama tuberculata Gray, 1827, Zool. Jour., vol. 3, p. 218 (type locality: "Bengal"; holotype 
in British Museum) . 

Distribution. The western Himalayas from Chitral and Kashmir through 
the Alpine Punjab to Katmandu District in Nepal. Smith's (1935, pp. 214-216) 
record for Kabul is the only documented Afghan occurrence, and this record 
may refer to A. nuristanica Anderson and Leviton. 

Genus Calotes Cuvier 

Calotes Cuvier, 1817, Regne Anim., vol. 2, p. 35 (type species: Lacerta calotes Linnaeus, 
1758, by absolute tautonomy). 

Calotes versicolor (Daudin). 

Agama versicolor Daudin, 1802, Hist. nat. Rept., vol. 3, p. 395, pi. 44 (type locality: 
Pondicherry, India [restricted by Smith, 1935, Fauna British India, vol. 2, p. 192]; 
holotype in Paris Museum). 

Calotes versicolor, Gray, 1845, Cat. spec. liz. British Mus., p. 243. 

Distribution. The entire Indian and Indo-Chinese subregions; southeastern 
Afghanistan; extreme eastern Iran; Pakistan; the entire Indian Peninsula; 



Vol. XXXVIII] LEVITOX & AXDERSON: AFGHANISTAN HERPETOFAUNA 179 

Nepal; Ceylon; Andaman Islands; Pulo Condore; Hainan; Hong Kong; 
southern China; the northern part of the Malay Peninsula; Sumatra; in West 
Pakistan it ranges north to Swat; locally distributed in Baluchistan. Only two 
documented localities for Afghanistan, both in the eastern part of the country, 
south of the Hindu Kush, the highest at 7400 feet elevation. 

Genus Phrynocephalus Kaup 

Phrynocephaltis Kaup, 1825, Isis von Oken, vol. 1, p. 591 (type species: Lacerta guttata 
Gmelin, 1789, by subsequent designation by Fitzinger, 1843, Syst. Rept., pp. 18 and 88) 
[ICZN, 1964: 69a, iv]. 

Phrynocephalus clarkorum Anderson and Leviton. 

Phrynocephalus ornatus Boulenger (in part), 1887, Cat. liz. British Mus., vol. 3, pp. 496-497. 

Phrynocephalus clarkorum Anderson and Leviton, 1967, Proc. California Acad. Sci., ser. 4, 
vol. 35, pp. 228-231, fig. 1 (type locality: southeast of Kandahar, .^Afghanistan, 31°20' 
N., 6S°50' E.; holotype: California .A.cademy of Sciences no. 97989). 

Distribution. The Helmand River basin of Afghanistan and West Pakistan. 

Phrynocephalus euptilopus Alcock and Finn. 

Phrynocephalus euptilopus Alcock and Finn, 1896, Jour. Asiatic Soc. Bengal, vol. 65, pt. 2, 
p. 556, pi. 12 (type locality: West Pakistan near Darband [elev. 3000 feetl, Baluchistan; 
syntypes in Indian Museum, Calcutta; British Museum; Museum of Comparative 
Zoology, Harvard [MCZ 7227$]). 

Distribution. Known from the six syntypes which were collected near 
Darband, West Pakistan, 3000 feet elevation, a small hollow in the Baluchistan 
desert basin region at the Afghan frontier, and from recently collected specimens 
from southeast of Darweshan, in the Helmand basin, Afghanistan (California 
Academy of Sciences nos. 120205-120207). 

Phrynocephalus interscapularis Lichtenstein. 

Phrynocephalus interscapularis Lichtenstein, 1856, Nomen. Rept. Amphib. Mus. Zool. Berol., 
p. 12 (type locality: Bokhara, USSR). 

Distribution. Central Asian republics of USSR, north to the southeastern 
list Urt, Aral Sea, and Aral Kara Kums, east to the valley of the Syr Darya 
River, the Kara-tau Ridge, spurs of the Tien Shan and Pamiro-Alai; along the 
valley of the P3'andzh River it penetrates east to the Vakhsh Valley. It is known 
in lowland northern Afghanistan from the area between Andkhui and IVIazar-i- 
Sharif, below 2000 feet elevation. 

Phrynocephalus luteoguttatus Boulenger. 

Phrynocephalus luteoguttatus Boulenger, 1887, Cat. liz. British Mus., vol. 3, p. 497 (type 
locality: between Nushki and Helmand, along Afghan-Baluch border; Helmand, Af- 
ghanistan; syntypes in British Museum). 

Distribution. Afghanistan in the Helmand River basin. West Pakistan in 
the desert basins of the Nushki and Chagai districts and western Las Bela. It 



180 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

probably enters Seistan in Iran, but no reliable Iranian records exist; to about 
4000 feet elevation. 

Phrynocephalus maculatus Anderson. 

Phyy)wcephalus maculatus John Anderson, 1872, Proc. Zool. Soc, London, p. 389 (type local- 
ity: Awada, Iran [corrected to Abadeh, north of Shiraz, by Blanford, 1876, Zool. E. Per- 
sia, vol. 2, p. 331]). 

Distribution. The Afghan-Baluchistan border region, east as far as Nushki; 
the Central Plateau of Iran; northern and gulf coastal Arabia, and Iraq. The 
extent of its penetration into Afghanistan is not known, though it does reach 
Darweshan in the Helmand basin; it occurs in Iranian Seistan, on the Central 
Plateau in Iran, and may be expected throughout the low southern and western 
regions of Afghanistan. 

Phrynocephalus mystaceus (Pallas). 

Lacerta mystacea Pallas, 1776, Reise versch. Prov. russ. Reich., vol. 3, p. 702, pi. 5, fig. 1 
(type locality: Naryn steppe on north coast of Caspian Sea, USSR [restricted by 
Mertens and Muller, 1928, Abh. Senckenberg. naturf. Ges., vol. 41, p. 26]). 

Phrynocephalus mystaceus, Boltlenger, 1885, Cat. liz. British Mus., vol. 1, pp. 379-380. 

Distribution. Found in Central Asian USSR, southern Kazakhstan; 
Astrakhan District, eastern Ciscaucasia, northeastern and eastern Iran, and 
adjacent regions of Afghanistan; in Ciscaucasia south to the vicinity of Ma- 
khach-kala; the western limit of distribution passes between the Volga and the 
Don, the northern limit reaches 48-50° N in some places (Irgiz, Turgai), and in 
eastern Kazakhstan to the Balkhash and Ala-kul lakes. To the east the distri- 
bution is limited by the foothills of Tien Shan and l^amiro-Alai; it somewhat 
surpasses the Termez along the valley of the Amu Darya (Terentjev and 
Chernov, 1949, p. 159). In Afghanistan it occurs north of the central massif; 
it may extend south of the Hari Rud along the western border inasmuch as it 
has been recorded from northeastern Iran (Zirkuch region) near the Afghan 
border. 

Phrynocephalus ornatus Boulenger. 

Phrynocephalus ornatus Boulenger (in part), 1887, Cat. liz. British Mus., pp. 496-497 (type 
locality: between Nushki and Helmand, Afghan-Baluch border; lectotype: British 
Museum no. 1946.8.28.20 [Anderson and Leviton, 1967, Proc. California Acad. Sci., ser. 
4, vol. 35, pp. 231-233]). 

Distribution. The Helmand River basin of Afghanistan and the desert 
basins of Baluchistan, West Pakistan. Also recorded from the Zirkuch region of 
eastern Iran (Nikolsky, 1897, p. 324). Whether these latter specimens belong 
to Phrynocephalus ornatus or to P. clarkorum, or both, has not been determined. 



Vol. XXXMIII LEVITOX & AXDERSOX: AFGHAXISTAX HERPETOFAUXA 181 

Phrynocephalus reticulatus boettgeri Bedriaga. 

Phrynocephalus raddei var. boettgeri Bedrtaca, 1907, Wissenschaft Result. N. M. Prewalski 

Central Asien Reisen, vol. 3, p. 217 (type locality: Shirabad, Uzbekistan, USSR; holo- 

type: Zoological Institute Leningrad no. 6117). 
Phrynocephalus reticulatus boettgeri, Terentjev and Chernov, 1949, Diag. Rept. Amph., 

p. 154. 

Distribution. Southwestern Tajikistan and adjacent regions of Uzbekistan; 
northern lowland Afghanistan. 

Phrynocephalus scutellatus (Olivier). 

Agama scutellata Olivier, 1807, Voy. Emp. Otho. (ed. 4), vol. i, p. 110, Atlas, pi. 42, fig. 1 

[not seen] (type locality: Mt. Sophia, near Isfahan, Iran; holotype: Paris Museum 

no. 6947). 
Phrynocephalus tickelii Gr.ay, 1845, Cat. Hz. British Mus., p. 260 (type locality: Afghanistan; 

holotype in British Museum). 
Phrynocephalus scutellatus, Smith, 1935, Fauna British India, vol. 2, Sauria, p. 229. 

Distribution. The Central Plateau of Iran, southern Iran, southern Afghan- 
istan, and northern Baluchistan, West Pakistan; to at least 7200 feet elevation 
in Afghanistan. 

Genus Uromastyx Merrem 

Uromastyx Merrem, 1820, Tent. Syst. Amph., p. 56 (type species: Stellio spinipes Daudin, 
1802, by subsequent designation by Fitzinger, 1843, Syst. Rept., p. 18). 

Uromastyx asmussi (Strauch). 

Centrotrachelus asmussi Strauch, 1863, Bull. Acad. Imp. Sci. St. Petersbourg, vol. 6, col. 479 

(type locality: Sar-i-Tschah, Iran; holotype in Zoological Institute Leningrad). 
Uromastix asmussi, Boulenger, 1885, Cat. liz. British Mus., vol. 1, p. 409. 

Distribution. The eastern portion of the Central Plateau of Iran, and the 
adjacent areas of West Pakistan and Afghanistan; 1800-4000 feet elevation in 
Iran. The extent of its occurrence in Afghanistan is unknown, as there exists a 
single record for that country, on the Afghan-Baluch border. 

Uromastyx hardwickii Gray. 

Uromastix hardwickii Gray, 182 7, Zool. Jour., vol. 3, p. 219 (type locality: Kanauj district, 
United Provinces, India; type in British Museum). 

Distribution. India, from the United Provinces to Kathiawar and west to 
the Northwest Frontier Provinces and southeastern Baluchistan east of the Ira- 
nian Plateau and below 2000 feet elevation. Afghanistan along the Kabul River 
Valley. 

Family Anguidae 
Genus Ophisaurus Daudin 

Ophisaurus D.audin, 1803, Bull. Soc. Phil., vol. 3, p. 188 (type species: Anguis ventralis 
Linnaeus, 1758, by monotypy). 



182 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser 

Ophisaurus apodus (Pallas). 

Lacerta apoda Pallas, 1775, Nov. Comment. Acad. Sci. Petropol., vol. 19, p. 435, pis. 9-10 

[not seen] (type locality: Naryn steppe on the north coast of the Caspian Sea, USSR). 

Ophisaurus apodus, Mertens and Muller, 1928, Abh. Senckenberg. naturf. Ges., vol. 41, p. 26. 

Distribution. Balkan Peninsula (to Istria and southern Dobrudsha in the 
north); Crimean Peninsula; Turkey; Rhodes; Syria; northern and v^estern 
Iran; Caucasus, Transcaspian, and Turkestan regions of USSR, and northern 
Afghanistan north of the Hindu Rush; to at least 8700 feet in Afghanistan. 

Family Gekkonidae 
Genus Agamura Blanford 

Agamura Blanford, 1874, Ann. Mag. Nat. Hist., ser. 4, vol. L% p. 455 (type species: 
Gymnodactylns persiciis Dumeril, 1856, by subsequent designation by Smith, 1935, Fauna 
British India, vol. 2, p. 61). 

Agamura femoralis: See Addendum, pg. 205. 

Agamura persica (Dumeril). 

Gymnodactylns persicus Dumeril, 1856, Arch. Mus. Hist. Nat. Paris, vol. 8, p. 481 (type 

locality: Iran; syntypes: Paris Museum no. 6761 |.5l). 
Agamura persica, Blanford, 1874, Ann. Mag. Nat. Hist., ser. 4, vol. 13, p. 455. 

Distribution. Iran, on the Central Plateau; West Pakistan, east to Cape 
Monze near Karachi and inland to Waziristan ; Afghanistan in the areas south of 
the Hindu Kush and the low country west of the mountains, along the Iranian 
border; to 8500 feet elevation at Paghman. 

Alsophylax pipiens: See Addendum, pg. 205. 

Genus Bunopus Blanford 

Bunopus Blanford, 1874, Ann. Mag. Nat. Hist., ser. 4, vol. 13, p. 454 (type species: Bunopus 
tuberculatus Blanford, 1874, by monotypy). 

Bunopus tuberculatus Blanford. 

Bunopus tuberculatus Blanford, 1874, Ann. Mag. Nat. Hist., ser. 4, vol. 13, p. 454 (syntypes 
from Iran: Bahu Kalat; Pishin; Isfandak; near Bampur; Rigan ; Narmashir: Tunb 
Island; West Pakistan: Baluchistan: Mand; Saman; Dasht; syntypes: British Mu- 
seum; Indian Museum, Calcutta; Museum of Comparative Zoology, Harvard no. 7128). 

Distribution. Iraq; Iran; Afghanistan; West Pakistan. All Afghan records 
are for Seistan, the Helmand Basin, and the higher regions to the east (7000 
feet elevation at Ghaomi Faringi). 

Genus Crossobamon Boettger 

Crossobamon Boettger, 1888, Zool. Jahrb. HI, Syst., p. 880 (type species: Gymnodactylns 
eversmanni Wiegmann, 1834, by monotypy). 

Crossobamon eversmanni (Wiegmann). 

Gymnodactylns eversmanni Wiegmann, 1834, Herpet. Mexicana, p. 19, note 28 (type locality: 

"Asia media") . 
Crossobamon eversmanni, Boettger, 1888, Zocl. Jahrb. Ill, Syst., p. 880. 



Vol. XXXVIII] LEVITOX & AXDERSOX: AFGHAMSTAX HERPETOFAUNA 183 

Distribution. Central Asian republics of the USSR, northeastern and east- 
ern Iran and neighboring regions of Afghanistan. In the north it ranges to the 
northern Chink (Precipice), Ust Urt, Irgiz River, Aral Kara Kum, and sands of 
Muyun Kum, east to the mountain system of Tien Shan and Pamiro-Alai 
(Terentjev and Chernov, 1949, p. 129). 

Crossobamon lumsdeni (Boulenger). 

Stenodactylns lumsdeni Boulenger, 1887, Cat. liz. British Mus., vol. 3, p. 479 (type locality: 

Afghan-Baluch border between Nushki and Helmand; holotype in British Museum). 
Crossobamon lumsdeni, Kluge, 1967, Bull. .American Mus. Nat. Hist., vol. 135, p. 23. 

Distribution. Known from the holotype, which was taken somewhere be- 
tween Nushki, West Pakistan, and the Helmand River, Afghanistan. Xikolsky 
(1899, p. 388) records it from Gurmuck, eastern Kerman, Iran. 

Crossobamon maynardi (Smith). 

Stenodactylns orientalis, Alcock and Finn, 1896 {ner Bl.\nford, 1876), Jour. Asiatic Soc. 
Bengal, vol. 65, p. 554. 

Stenodactylns maynardi Smith, 1933, Rec. Indian Mus., vol. 35, p. 18 (type locality: 
Baluchistan, near Afghan border; based on Alcock and Finn's specimens; syntypes: 
British Museum no. 1931.6.14.1 9, and Indian Museum, Calcutta, no. 13944 c5)- 

Crossobamon maynardi, Kluge, 1967, Bull. American Mus. Nat. Hist., vol. 135, p. 23. 

Distribution. The types were collected somewhere along the route followed 
by Maynard and McMahon during the travels of the Afghan-Baluch Boundary 
Commi.ssion of 1896. No definite locality was recorded. Minton (1966, p. 165) 
records specimens from the vicinity of Nushki, Chagai District, Baluchistan, 
West Pakistan. It is known from the Helmand Basin in Afghanistan. 

Genus Cyitodactylus Gray 

Cyrtodactylus Gray, 1827, Phil. Mag., ser. 2, vol. 2, p. 55 (type species: Cyrtodactylus 
pulchellus Gray, 1827, by monotypy). 

Cyrtodactylus caspius ( Eichwald ) . 

Gymnodactyliis caspius Eichwald, 1831, Zool. Spec, vol. 3, p. 181 (type locality: Baku, on 
the Caspian Sea, USSR; syntypes: Zoological Institute Leningrad nos. 3181-3182 [?]). 
Cyrtodactylus caspius, Underwood, 1954, Proc. Zool. Soc. London, vol. 124, p. 475. 

Distribution. Eastern Azerbaidzhan SSR, northern and eastern Iran, 
northern Afghanistan, and Central Asian republics of the USSR, north to a line 
connecting the southeastern coast of the Kosomolets Gulf with the northern coast 
of the Aral Sea, and east to the mountains in the Kysyl Kums, the Nura Tau 
Ridge, and foothills of the Pamiro-Alai; along the valley of the Pyandzh River 
it reaches east to the surroundings of Chubek (Terentjev and Chernov, 1949, p. 
137). In Afghanistan it occurs north of the Hindu Kush. 



184 C ALIFORM A ACADEMY OF SCIEXCES LProc. 4th Ser. 

Cyi todactylus fedtschenkoi (Strauch). 

Gymnodactylus fedtschenkoi Strauch, 1887, Mem. Acad. Imp. Sci. St. Petersbourg, ser. 7, 
vol. 35, pp. 46-47 (syntypes from USSR: Samarkand; Bokhara; Gissar; syntypes: 
Zoological Institute Leningrad nos. 338712], 5039[2], 6354, 535S[4], 6479, 7401[2]). 

Cyrtodactylus fedtschenkoi, Underwood, 1954, Proc. Zool. Soc. London, vol. 124, p. 475. 

Distribution. Western Turkmen, Uzbekistan, Tajikistan, Afghanistan, 
eastern Iran, and western Baluchistan. In the USSR the westernmost locaHties 
are the valley of the Tedzhen River, surroundings of Mara, coast of the Aral Sea ; 
the northeastern border of the range does not reach beyond the valley of the 
Syr Darya River. In Afghanistan it is known from north of the Hindu Rush 
and has been found at Paghman. It has been suggested that Gymnodactylus 
longipes Nikolsky of eastern Iran is a synonym (Clark, Clark, Anderson, and 
Leviton, 1969, p. 301). 

''Cyrtodactylus russowii (Strauch). 

Gymnodactylus russowii Strauch, 1887, Mem. Acad. Imp. Sci. St. Petersbourg, ser. 7, vol. 

35, pp. 49-51, figs. 10-12 (syntypes from USSR: Novo-Alexandrovsk ; Chodschent; 

Mangyschlak; Murza Robat; Mohol-tau ; Tschimkent; Tschinaz ; Golodnaja desert; 

Utsch-Kurgan at Naryn; Clark-Ukjur; syntypes: Zoological Institute Leningrad nos. 

3658[2], 3659, 3660, 3700[3], 3701[2], 4192, 4193r6], 4194, 4195[5], 4310[2], 5037, 5197, 

5218, 5224, S800[2], California Academy of Sciences nos. 94050-94052). 
Cyrtodactylus russowii, Underwood, 1954, Proc. Zool. Soc. London, vol. 124, p. 475. 

Distribution. Eastern Caucasus foreland (Starogladkowskaja) and from 
the east coast of the Caspian Sea to central Asia, including northern Iran and 
northern Afghanistan, according to Wermuth (1965, p. 66). We know of no 
documented records for either Iran or Afghanistan. 

Cyrtodactylus scaber (Heyden). 

Stenodactylus scaber Heyden, 182 7, in Rxjppell, Atlas Reise nordl. Afrika, Rept., p. 15, pi. 
4, fig. 2 (type locality: Tor, Sinai Peninsula [see Anderson, J., 1898, Zool. Egypt, p. 55, 
for comment on probable source of Heyden's specimens] ; lectotype: Senckenberg Museum 
Frankfurt no. 8180 $ ). 

Cyrtodactylus scaber, Underwood, 1954, Proc. Zool. Soc. London, vol. 124, p. 475. 

Distribution. From Egypt south to Ethiopia, and east across Arabia and 
the arid regions of Southwest Asia to Afghanistan, West Pakistan, and north- 
western India. In Afghanistan it is known from the low elevations of the south 
and southeast, up to about 5000 feet elevation. Our previous (Clark, Clark, 
Anderson, and Leviton, 1969, p. 302) identification of a small juvenile gecko 
from northwestern Afghanistan is in error. This specimen has been reidentified 
as C. caspius. 

Cyrtodactylus watsoni (Murray). 

Gymnodactylus watsoni Murray, 1892, Zool. Beloochistan and S. Afghanistan, pp. 68-69 

(type locality: Quetta, West Pakistan). 
Cyrtodactylus watsoni, Minton, 1966, Bull. American Mus. Nat. Hist., vol. 134, p. 79. 



Vol. XXXVIII] LEVITOX & ANDERSON: AFGHANISTAN HERPETOFAUNA 185 

Distribution. Northern Las Bela to Quetta, and northeastward to Swat 
and the northern Punjab in West Pakistan; westward up to the Kabul River 
Valley at least as far as Jalalabad in Afghanistan. 

Cyrtodactylus species. 

Cyrtodactylits jedtschenkoi, Clark, Clark, Anderson, and Leviton, 1969, Proc. California 

Acad. Sci., ser. 4, vol. 36, pp. 300-302 [but not including fig. 2]. Anderson and Leviton, 

1969, Proc. California Acad. Sci., ser. 4, vol. 37, p. 45. 

Distribution. A Cyrtodactylus closely allied to C. jedtschenkoi occurs in 
the Helmand Basin from Farah in the west to near Kandahar in the east. This 
appears to be an undescribed species, and questions regarding the systematics of 
these and related geckos are under study. 

Genus Eublepharis Gray 

Eiiblepharis Gray, 182 7, Phil. Mag., ser. 2, vol. 2, p. 56 (type species: Eublepharis hardwkkii 
Gray, 1827, by monotypy). 

Eublepharis macularius (Blyth). 

Cyrtodactylus macularius Blyth, 1854, Jour. Asiatic Soc. Bengal, vol. 23, pp. 737-738 (type 

locality: Salt Range, Punjab; holotype in Indian Museum, Calcutta). 
Eublepharis macularius, John Anderson, 1871, Proc. Zool. Soc. London, p. 163. 

Distribution. Southern Transcaspia, eastern Afghanistan south of the 
Hindu Kush, West Pakistan in Baluchistan, the Northwest Frontier Provinces, 
and south to Rajputana and the Khandesh District of India; to at least 5300 
feet elevation in Afghanistan, 8000 feet in Baluchistan. As yet, specimens are 
unknown from the large area between southern Turkmen and eastern Afghani- 
stan. 

Genus Hemidactylus Oken 

Hemidactylus Oken, 1817, Isis von Oken, col. 1183 (based on Cuvier's Hemidactyle, 1817, 
Regne Anim., vol. 2, p. 47; type species: Gecko tuberculosus Daudin). 

Hemidactylus flaviviridis Riippell. 

Hemidactylus flaviviridis Ruppell, 1835, Neue Wirbelth. Faun. Abyss., Amph., p. 18, pi. 6, 

fig. 2 (type locality: Nassaua Island, Eritrea; lectotype: Senckenberg Museum Frankfurt 

no. 8772 cJ). 

Distribution. Northern India west of Bengal and south to Bombay, through 
southern (coastal) Iran and Arabia to the African shores of the Red Sea. In 
Afghanistan it is known only from Paghman and Jalalabad. Much of its distribu- 
tion, from the shores of the Red Sea and around the shores of the Arabian 
Peninsula and Iran, is due to its having been carried about by man, and its 
presence in Afghanistan may also be due to human agency. 

Genus Teratoscincus Strauch 

Teratoscincus Strauch, 1863, Bull. Acad. Imp. Sci. St. Petersbourg, vol. 6, col. 480 (type 
species: Teratoscincus keyserlingi Strauch, 1863, by monot\py). 



186 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

Teratoscincus bedriagai Nikolsky. 

Teratoscinciis bedriagai Nikolsky, 1899, Ann. Mus. Zool. Acad. Imp. Sci., St. Petersbourg, 
vol. 4, pp. 146-147 (types from Seistan and Zirkuch, eastern Iran; syntypes: Zoological 
Institute Leningrad nos. 9157, 91S8[2], 9159[3], 9160, 9161, 9162, 9163). 

Distribution. Eastern Iran and the Helmand River basin of Afghanistan; 
to at least 4700 feet elevation. 

Teratoscincus microlepis Nikolsky. 

Ceramodactylus af finis, Alcock and Finn, 1896 (nee Murray, 1884), Jour. Asiatic Soc. 

Bengal, vol. 65, p. SS4. 
Teratoscineus microlepis Nikolsky, 1899, Ann. Mus. Zool. Acad. Imp. Sci., St. Petersbourg, 

vol. 4, pp. 145-146 (type locality: Duz-Ab ; holotype: Zoological Institute Leningrad 

no. 9164). 

Distribution. Extreme eastern Iran and adjacent Baluchistan and Afghani- 
stan along the Afghan-Baluchistan border. 

Teratoscincus scincus (Schlegel). 

Stenodaetylus scincus Schlegel, 1858, Handl. Dierk., vol. 2, p. 16 (type locality: Hi River, 

Turkestan, USSR; holotype in Leiden Museum). 
Teratoscincus scincus, Boulenger, 1885, Cat. liz. British Mus., vol. 1, pp. 12-13, pi. 2, fig. 3. 

Distribution. Central Asia, in the north up to the Chink (Precipice) Ust 
Urt, Aral Kara Kum, and valleys of the rivers Chu and Hi, east to the foothills 
of the Tien Shan and Pamiro-Alai ; one record for Sachow in the southern Gobi ; 
the vicinity of Kokand in the Syr Darya Valley and Vakhsh Valley inclusive in 
the valleys of the river Pyandzh; northeastern and eastern Iran (Terentjev and 
Chernov, 1949, p. 128). To the west it reaches the eastern shore of the Caspian 
Sea. Its western Hmit in Iran is the steppe between Argavani and Marinjab, 
Tehran Province. In Baluchistan it is not known east of Nushki, nor south of 
Kharan (Minton, 1966, p. 76). It is known in Afghanistan from the low deserts 
along the western and southern borders; to 6000 feet elevation in Iran, at least 
4700 feet in Afghanistan. 

Family Lacertidae 
Genus Acanthodactylus Fitzinger 

Acanthodactylus Fitzinger, 1834, in Wiegmann, Herpet. Mexicana, p. 10 (type species: 
Lacerta boskiana Lichtenstein, 1823, by monotypy). 

Acanthodactylus cantoris Giinther. 

Acanthodactylus cantoris Gunther, 1864, Rept. British India, p. 73 (type locality: Ramnagar, 
Agra, India; holotype in British Museum). 

Distribution. The species as a whole ranges from northwestern India 
through West Pakistan, southern Afghanistan, and lowland southern Iran and 
Arabia. In Afghanistan it occurs in the valley of the Kabul River and in the 
Helmand River basin. The status of these populations is under study. 



Vol. XXXVIII] LEVITON & ANDERSON: AFGHANISTAN HERPETOFAUNA 187 

Genus Eremias Fitzinger 

Eremias Fitzinger, 1834, in Wiegmanx, Herpet. Mexicana, p. 9 (type species: Lacerta 
variabilis Pallas, 1811, by subsequent designation by Fitzinger, 1843, Syst. Rept., p. 21). 

Eremias acutirostris (Boulenger). 

Scapteira acutirostris Boulenger, 1887, Cat. liz. British Mus., vol. 3, pp. 114-115 (type 

locality: between Nushki and Helmand, Afghan-Baluch border region; holotype in 

British Museum). 
Eremias {Scapteira) acutirostris, Lantz, 1928, Bull. Mus. Georgie, vols. 4 and 5, pp. 41, 136. 

Distribution. Desert basins of northwestern Baluchistan and adjoining 
Afghanistan. 

Eremias aporosceles (Alcock and Finn). 

Scapteira aporosceles Alcock and Finn, 1896, Jour. Asiatic Soc. Bengal, vol. 65, p. 559, pi. 13 
(type locality: Afghan-Baluch border: "common west of Robat I" frestricted by Smith, 
1935, Fauna British India, vol. 2, p. 388, to Baluchistan: near Nushki; Robat I lies some 
120 miles to the west of Nushki, however] ; syntypes in British Museum, and Indian 
Museum, Calcutta). 

Eremias {Scapteira) aporosceles, Lantz, 1928, Bull. Mus. Georgie, vols. 4 and 5, pp. 41, 127- 
130, 136. 

Distribution. Baluchistan, West Pakistan, and ^Afghanistan, along the 
Afghan-Baluch border. 

Eremias aria Anderson and Leviton. 

Eremias aria Anderson and Leviton, 1967, Occ. Pap. California Acad. Sci., no. 64, pp. 1-4, 
fig. 1 (type locality: 5-10 mi. ENE. Nimla on old Kabul-Jalalabad road, 10 mi. SW. 
Balabagh [34°19-21' N, 70°10-15' E] ; holotype: California Academy of Sciences no. 
96204 $). 

Distribution. Known only from the vale of Jalalabad in eastern Afghan- 
istan. 

Eremias fasciata Blanford. 

Eremias fasciata Blanford, 1874, Ann. Mag. Nat. Hist., ser. 4, vol. 14, p. 32 (type locality: 
Saidabad, southwest of Kerman, Iran; syntypes in British Museum). 

Distribution. Eastern Iran, southern Afghanistan in the Helmand River 
basin, and Baluchistan, West Pakistan. 

Eremias grammica (Lichtenstein). 

Lacerta grammica Lichtenstein, 1823, in Eversmann, Reise nach Buchara, p. 140 (type 

locality: Karakum and Kizyl-Kum, USSR). 
Eremias {Scapteira) grammica, Lantz, 1928, Bull. Mus. Georgie, vols. 4 and 5, pp. 41, 

117-122, 136. 

Distribution. Central Asian republics of the USSR, southern Kazakhstan, 
north to lower reaches of Irghiz and Turgai rivers and Lepsa River, east to Ala 
Tau mountains, northeastern and eastern Iran, and adjacent lowland regions of 
Afghanistan, north of the Hindu Kush. 



188 CALIFORNIA ACADEMY OF SCIENCES LProc. 4th Ser. 

Eremias guttulata watsonana Stoliczka. 

Eremias (Mesalhia) watsonana Stoliczka, 1872, Proc. Asiatic Soc. Bengal, pp. 86-87 (type 

locality: between Karachi and Sakhar, Sind, West Pakistan; syntypes: British Museum, 

and Indian Museum, Calcutta). 
Eremias guttulata watsonana, Smith, 1935, Fauna British India, vol. 2, pp. 389-390. 

Distribution. Eremias guttulata ranges from North Africa through Arabia 
and the desert regions of Southwest Asia, north to Turkman and east to Sind 
in West Pakistan. E. g. watsonana occurs throughout Iran and Afghanistan at 
elevations below 8000 feet. According to Minton (1966, p. 110), it is found 
throughout the arid parts of West Pakistan but often rather spottily, common in 
Las Bela and along the edge of the Thar Desert, but rare in the intervening area. 

Eremias intermedia (Strauch). 

Podarces (Eremias) intermedia Strauch, 1876, Voy. Przewalski, Rept., p. 28 (type locality: 

Kizil Kum, Aralo-Caspian desert, USSR). 
Eremias intermedia, Boulenger, 1887, Cat. liz. British Mus., vol. 3, pp. 100-101. 

Distribution. Soviet Central Asia and southern regions of Kazakhstan; 
north to Mangyshlak, sands of the Bol'shie Barsuki, Aral Kara Kums, valley 
of Chu River and Balkhash Lake; east to Tien Shan and Pamiro-Alai; reaching 
the sands of the Vakhsh lowlands along the valley of the Amu Darya River. It 
has been taken in the valley of the Tajan River at the point where the borders of 
Iran, Afghanistan, and Turkmen meet, and undoubtedly occurs within the bor- 
ders of Iran and Afghanistan, although no records exist to the south of this point. 

Eremias lineolata (Nikolsky). 

Scapteira lineolata Nikolsky, 1896, Ann. Mus. Zool. Acad. Imp. Sci., St. Petersbourg, vol. 1, 
p. 371 (type locality: between Faizabad and Xusi, eastern Iran; syntypes: Zoological 
Institute Leningrad no. 8801|6l ; British Museum). 

Eremias lineolata, Lantz, 1928, Bull. Mus. Georgie, vols. 4 and 5, pp. 39, 79-84, 134. 

Distribution. Turkmen, Lizbekistan, southern Kazakhstan, southwestern 
Tajikistan, USSR. It occurs in eastern Iran and northern lowland Afghanistan. 
In the north it ranges up to the Chink (Precipice) Ust Urt, Aral Sea, middle 
and lower course of the Chu River and lower coast of Balkhash Lake, east to the 
Tien Shan and Pamiro-Alai mountain system. To the east it reaches the lower 
extent of the Vakhsh River inclusive along the Amu Darya Valley. 

Eremias nigrocellata Nikolsky. 

Eremias nigrocellata Nikolsky, 1896, Ann. Mus. Zool. Acad. Imp. Sci. St. Petersbourg, vol. 
1, p. 371 (types from between Feizabad and Mondechi, and Seistan, eastern Iran; syn- 
types: Zoological Institute Leningrad nos. 8798r3], 8779|2l, 8800). 

Distribution. Southwestern Tajikistan, southern Uzbekistan (vicinity of 
Shirabad) in the USSR; eastern Iran; northern lowland Afghanistan. It occurs 
between 4000 and 5000 feet elevation in Iran; known Afghan localities are below 
2000 feet. 



Vol. XXXVIIIl LEVITOX & AXDERSOX: AFGHAXISTAN HERPETOFAUXA 189 

Eremias regeli Bedriaga. 

Eremias regeli Bedriaga, 1907, Ann. Zool. Mus. Acad. Imp. Sci. St. Petersbourg, vol. 10 
(1905), p. 236 (type locality: Shirabad, Uzbekistan, USSR; syntype: Zoological In- 
stitute Leningrad no. 6115). 

Distribution. Found in the USSR in the valleys of the upper reaches of the 
Amu Darya River and lower course of the Pyandzh River and their tributaries, 
and adjacent foothills. In the west it is known up to the vicinity of Kelif, east 
of Kulyab, north to the Gissar Ridge (Terentjev and Chernov, 1949, p. 199). 
The only Afghan record is in the valley of the Kabul River, to the south of the 
Hindu Rush, an unexpected occurrence suggesting serious unsolved systematic 
and zoogeographic problems. 

Eremias scripta (Strauch). 

Podarces (Scapteira) scripta Str.-^uch, 1867, Mel. Biol. Acad. St. Petersbourg, vol. 6, p. 424 

(type locality: Aralo-Caspian desert, USSR; no specimens listed, nor type designated). 
Eremias (Rhabderemias) scripta, Lantz, 1928, Bull. Mus. Georgie, vols. 4 and 5, pp. 38, 

73-79, 133. 

Distribution. Soviet Central Asia and southern Kazakhstan; in the north 
to Mangyshlak, southern Ust Urt Precipice, sands of Bol'shie Barsuki, valley 
of Chu River, coast lines of Lake Balkhash, and valley of the Lepsa 
River, east to the Tien Shan and Pamiro-Alai mountain ranges; along the 
Amu Darya Valley it reaches the lowlands of the Vakhsh River. Terentjev 
and Chernov (1949, p. 204) state that it is found in eastern Iran and 
adjacent regions of Afghanistan and Baluchistan. We find no records for Iran; 
in Afghanistan it occurs in the southern desert region, and in Baluchistan it is 
known from the Chagai District. If this southern population is, indeed, the same 
species as that inhabiting the USSR, it is to be expected along the Afghan- 
Iranian border. 

Eremias velox persica Blanford. 

Eremias persica Blanford, 1874, Ann. Mag. Nat. Hist., ser. 4, vol. 14, p. 31 (type locality: 

near Isfahan, Iran; syntypes in British Museum). 
Eremias velox var. persica, Boulenger, 1921, Monogr. Lacert., vol. 2, pp. 312-314. 

Distribution. The Central Plateau of Iran, southern Turkmen (vicinity of 
Kushka and Kopet Dagh), southern Afghanistan, and Baluchistan and Waziri- 
stan. West Pakistan. To at least 8000 feet in Afghanistan. 

Eremias velox velox (Pallas). 

Lacerta velox Pallas, 1771, Reise Russ. Reich, vol. 1, p. 457 (type locality: Inderskija 

Gory, region of lower Ural River, USSR). 
Eremias velox, Wiegmann, 1834, Herpet. Mexicana, p. 9. 
Eremias velox velox, Lantz, 1918, Proc. Zool. Soc. London, p. 14. 

Distribution. From the Volga to western Mongolia and into Sinkiang. In 
the southeast it is limited bv the Tien Shan Mountains, and in the southwest by 



190 CALIFORNIA ACADEMY OF SCIENCES IProc. 4th Ser. 

the Elburz Mountains and the south coastal region of the Caspian Sea. The 
Kopet Dagh forms its southern Hmit, except where it may penetrate the north- 
eastern border of Iran, extending also into northwestern Afghanistan, perhaps 
south along the Iran-Afghan border to the Seistan Basin. The only published 
records for Afghanistan are the River Tajan just at the Afghan-Iran-Trans- 
caspian border, and New Gulran in northwestern Afghanistan. 

Genus Ophisops Menetries 

Ophhops Menetries, 1832, Cat. rais. Obj. Zool. Caucas., p. 63 (type species: Ophisops 
elegans Menetries, 1832, by monotypy). 

Ophisops jerdoni Blyth. 

Ophisops jerdoni Blyth, 1853, Jour. Asiat. Soc. Bengal, vol. 22, p. 653 (type locality: 
Mhow, Indore, Central India; holotype lost [fide Smith, 1935, Fauna British India, 
vol. 2, p. 377]). 

Distribution. From the Kabul River Valley of eastern Afghanistan through 
the Northwest Frontier Provinces and northern Punjab of West Pakistan south 
to Rewa State and Bellary in western India. 

Family Scincidae 
Genus Ablepharus Fitzinger 

Ablepharus Fitzinger, 1823, in Lichtenstein, Ver. Doub. Zool. Mus. Berlin, p. 103 (type 
species: Ablepharus pannoniciis Lichtenstein, 1823, by monotypy). 

Ablepharus bivittatus lindbergi Wettstein. 

Ablepharus bivattatus lindbergi Wettstein, 1960, Zool. Anz., vol. 165, pp. 61-62 (type 
locality: steppe a few km. west of Obeh, east of Herat, northwestern Afghanistan; holo- 
type: Vienna Museum no. 15877). 

Distribution. Ablepharus bivittatus ranges from the Caucasus and Talysh 
mountains in southeastern Transcaucasia, USSR, northern Iran (to 11,000 feet 
elevation) and the Zagros Mountains of western Iran, through southern Turkmen 
in the Kopet Dagh, to Afghanistan and the Punjab. Ablepharus b. lindbergi is 
known from upland Afghanistan (to at least 9600 feet elevation), and a single 
record from the Punjab. 

* Ablepharus grayanus (Stoliczka). 

Blepharosteres grayanus Stoliczka, 1872, Proc. Asiat. Soc. Bengal, pp. 74-75 (type locality: 
Waggur District, northeastern Kachh, West Pakistan; holotype in Indian Museum, 
Calcutta). 

Ablepharus grayanus, Boulenger, 1887, Cat. Hz. British Mus., vol. 3, p. 352. 

Distribution. In West Pakistan it is known from Kutch, Sind, mostly west 
of the Indus, eastern Baluchistan, and the Punjab and Northwest Frontier Prov- 
inces at low elevations (Minton, 1966, p. 104). It is recorded from the eastern 



Vol. XXXVIII] LEVITON & ANDERSON: AFGHANISTAN HERPETOFAUNA 191 

and southeastern margins of the Central Plateau in Iran, and in the USSR from 
a single locality, Nimichi-Bol', in southern Tajikistan. There appear to be no 
documented records for Afghanistan, although Terentjev and Chernov (1949, p. 
171) state that it occurs there. 

Ablepharus pannonicus Lichtenstein. 

Ablephanis pannoniais Lichtenstein, 1823, in Eversmann, Reise nach Buchara, p. 145 
(type locality: Buchara, USSR). 

Distribution. Iraq; northern and western Iran; in the USSR it is found in 
the Kopet Dagh, mountains of the Pamiro-Alai systems and their foothill low- 
lands, north to the vicinity of Leninabad, east to Darvaz, inclusive; probably 
throughout most of Afghanistan to at least 7600 feet elevation; in West Pakistan 
it occurs in the mountainous northern sections from Quetta to Chitral ; northern 
Punjab. 

Genus Eumeces Wiegmann 

Eumeces Wiegmann, 1834, Herpet. Mexicana, p. 36 (type species: Scincus pavimentatus 
Geoffroy-St. Hillaire, 1827, by subsequent designation by Taylor, 1935, Univ. Kansas 
Sci. Bull., vol. 23, p. 29). 

*Eumeces blythianus (Anderson). 

Mabouia blythiana John Anderson, 1871, Proc. Asiat. Soc. Bengal, p. 186 (type locality: 

Amritzar, Punjab). 
Eumeces blythianus, Boulenger, 1887, Cat. liz. British Mus., vol. 3, p. 385. 

Distribution. Not yet recorded from within the political boundaries of 
Afghanistan, but known from the Afridi country along the Afghan border near 
the Khyber Pass. Minton (1966, p. 102) found it south to the coastal plain at 
Karachi. The type, said to come from Amritsar in the Punjab, was purchased 
from a merchant. 

Eumeces schneideri (Daudin). 

Scincus schneideri Daudin, 1802, Hist. nat. Rept., vol. 4, p. 291 (no type locaUty given). 
Eumeces schneideri, Boulenger, 1887, Cat. liz. British Mus., vol. 3, pp. 383-384. 

Distribution. Eumeces schneideri {sensu lata), is found across North Africa 
north of the Sahara; Southwest Asia, from the Mediterranean to West Pakistan, 
and from the Transcaucasian and Transcaspian provinces of the USSR to north- 
ern Saudi Arabia and the Persian Gulf. In Afghanistan it is known from the 
Helmand Basin and from northern Afghanistan in the low country north of the 
Hindu Kush. 

We have examined none of the specimens from Afghanistan. Terentjev and 
Chernov (1949, p. 169) state that Eumeces s. prince ps, the subspecies which 
occurs in the USSR, is also found in Afghanistan. Eumeces s. zarudnyi is the 



192 CALIFORNIA ACADEMY OF SCIENCES IProc. 4th Ser. 

form from eastern Iran. Both may occur in Afo;hanistan. Some recent workers 
regard E. blythianus as a subspecies of E. schneideri, and a record from the 
Helmand may refer to this form. We have not seen enough material to express 
an opinion regarding the relationships of the several nominal forms. 

Eumeces taeniolatus (Blyth). 

Eurylepis taeniolatus Blyth, 1854, Jour. Asiat. Soc. Bengal, vol. 23, pp. 739-740 (type 

locality: Salt Range, Punjab; holotype in Indian Museum, Calcutta). 
Eumeces taeniolatus, Stoliczka, 1872, Proc. Asiat. Soc. Bengal, pp. 75-76. 

Distribution. Southern Turkmen, USSR; eastern Iran; Afghanistan; West 
Pakistan; Kashmir; Arabia (two records exist for the Arabian peninsula, that 
of Taylor (1935), pp. 111-119) citing a specimen in the British Museum from 
El Kubar, southwestern Arabia, and Haas (1957, pp. 74-75, fig. 9) for a speci- 
men from 23 miles north of Hail, Saudi Arabia, and citing a British Museum 
specimen from Muscat). While Terentjev and Chernov (1949, p. 170) state 
that it is found in northern and northeastern Iran, the only record we find is 
that of the River Tajan on the Afghan-Iran-Turkmen border. In Afghanistan, 
the only known locality other than the River Tajan is Pandjvai near Kandahar. 

Genus Mabuya Fitzinger 

Mabuya Fitzinger, 1826, Neue Class. Rept., pp. 23 and 52 (type species: Lacerta mabouya 
Lacepede, 1788, by tautonomy). 

*Mabuya aurata (Linnaeus). 

Lacerta aurata Linnaeus, 1758 (in part), Syst. Nat., ed. 10, p. 209 (type locality: Cyprus). 
Mabuya aurata, Andersson, 1900, Kungi. Sv. Vet.-Akad. Handl. Stoclvhoim, vol. 26, pt. 4, 
p. 14. 

Distribution. Ethiopia; Eritrea; Cyprus; Syria; Turkey; Iracj; northern 
and western Iran; Muscat; Armenian SSR, Nakhichevan ASSR, southern Turk- 
men and Uzbekistan, north to Dzhizak and Chinaz. A record for Sind is in 
considerable doubt (Minton, 1966, p. 99), and it has not been recorded from the 
eastern part of the Plateau of Iran. The only possible Afghan record is for the 
River Tajan at the Iran-Afghanistan-Turkmen borders. 

Mabuya dissimilis (Hallowell). 

Euprepes dissimilis Hallowell, 1860, Trans. American Phil. Soc, vol. 11, p. 78 (type 

locality: Bengal). 
Mabuya dissimilis, Boulenc.er, 1887, Cat. liz. British Mus., vol. 3, p. 175. 

Distribution. From West Bengal and Bihar across the plains of northern 
and central India. In West Pakistan it ranges from the delta of the Indus north 
to Rawalpindi and Campbellpore, but not westward onto the Plateau of Iran 
(Minton, 1966, p. 101). In Afghanistan it is known only from the valley of the 
Kabul River, to 3500 feet elevation. 



Vol. XXXVIII] LEVITON & ANDERSON: AFGHANISTAN HERPETOFAUNA 193 

Genus Ophiomorus Dumeril and Bibron 

Ophiomonis Dumeril and Bibron, 1839, Erp. Gen., vol. 5, p. 799 (type species: Ophiomorus 
miliaris Dumeril and Bibron, 1839, by monotypy). 

Leviton's (1959, p. 461) inclusion of O. brevipes in the list of species known 
from Afghanistan was based on the speculation by Terentjev and Chernov (1949, 
p. 175) that it occurs in Afghanistan. 

Ophiomorus tridactylus (Blyth). 

Sphenocephalus tridactylus Blyth, 1855, Jour. Asiatic Soc. Bengal, vol. 22, p. 654 (type 

locality: Afghanistan; holotype in Indian Museum, Calcutta). 
Ophiomorus tridactylus, Boulenger, 1887 (in part). Cat. liz. British Mas., vol. 3, pp. 394- 

395. 

Distribution. The sandy areas of the Helmand Basin and adjacent regions 
of eastern Iran, southern Afghanistan, and northern Baluchistan, West Pakistan. 

Genus Scincella Mittleman 

Scincella Mittleman, 1950, Herpetologica, vol. 6, p. 19 (type species: Scincus lateralis 
Say, 1823, by original designation). 

Scincella himalayana (Giinther). 

Eumeces himalayanus Gunther, 1864, Rept. British India, p. 86 (type locality: Western 

Himalayas; holotype in British Museum). 
Scincella himalayana, Mittleman, 1950, Herpetologica, vol. 6, p. 19. 

Distribution. Mountainous regions from Nepal west to southern Turkmen; 
known from Chitral and the Hazara District of West Pakistan; Kashmir; Nuri- 
stan in eastern Afghanistan (based on two specimens in the Universitetets 
Zoologiske Museum from Pashki, identified as this species, but not seen by us). 

Family Varanidae 
Genus Varanus Merrem 

Varanus Merrem, 1820, Tent. Syst. Amph., p. 58 (type species: Lacerta varia Shaw, 1790, 
by subsequent designation of Gray, 1827, Phil. Mag., ser. 2, vol. 3, p. 55). 

Varanus bengalensis bengalensis (Daudin). 

Tupinambis bengalensis Daudin, 1802, Hist. nat. Rept., vol. 3, p. 67 (type locality: Bengal; 

holotype: Paris Museum no. 2179). 
Varanus bengalensis, Dumeril and Bibron, 1836, Erp. Gen., vol. 3, p. 480. 
Varanus (Indovaranus) bengalensis bengalensis, Mertens, 1942, Abh. Senckenberg. Naturf. 

Ges., no. 466, p. 334. 

Distribution. From southeastern Iran through West Pakistan and India to 
Assam and Burma, south to Tharawaddy and the Henzada District, north to 
Nepal, Bigrani (western Himalayas) and Darjeeling (eastern Himalayas). In 






yD^ 



'^J LIBRARY K^ 



194 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

Afghanistan it is known only from the valley of the Kabul River, to elevations 
of 8600 feet to the north of Jalalabad. 

Varanus griseus caspius (Eichwald). 

Psammosatirus caspius Eichwald, 1831, Zool. Spec, vol. 3, p. 190 (type locality: Dardsha 

Peninsula, east coast of Caspian Sea, USSR) . 
Varanus {Psanimosaurus) griseus caspius, Mertens, 1954, Senckenberg. Biol., vol. 35, p. 355. 

Distribution. The species ranges from North Africa through Southwest 
Asia to northern India. It reaches Rio de Oro in the west and Ambala, Agra, and 
Narsingarh in the east; northward it extends to the Transcaspian provinces of 
the USSR. Varanus g. caspius ranges from the eastern coast of the Caspian Sea 
through Central Asian republics of the USSR and southern Kazakhstan, where 
it is known up to the southern Ust Urt Precipice, coasts and islands of the Aral 
Sea, east to the Syr Darya Valley and mountains of the Tien Shan and Pamiro- 
Alai systems. It reaches east to Chubek along the valleys of the Amu Darya and 
Pyandzh rivers and is found throughout the Plateau of Iran, west to the Zagros 
Mountains, and southeast to northern Baluchistan. The area west of the Plateau 
of Iran is occupied by V. g. griseus, while V. g. koniecznyi is the form to the east. 
In Afghanistan the known locaUties are in the Helmand Basin in the southern 
part of the country, and the valley of the Hari-Rud in the north. It undoubtedly 
occurs in the lower elevations throughout Afghanistan. 

Suborder Serpentes 

Family Boidae 
Genus Eryx Daudin 

Eryx Daudin, 1803, Hist. nat. Rept., vol. 7, p. 251 (type species: Boa turcica Olivier, 1801, 
by subsequent designation by Fitzinger, 1843, Syst. Rept., p. 24). 

Eryx elegans (Gray). 

Cursoria elegans Gray, 1849, Cat. sn. British Mus., p. 107 (type locality: Afghanistan; 

holotype in British Museum). 
Eryx elegans, Blanford, 1876, Zool. E. Persia, vol. 2, p. 402. 

Distribution. Known from Paghman in Afghanistan. According to Terent- 
jev and Chernov (1949, pp. 229-230), Eryx jaculus czarcwskii Nikolsky, 1916, 
is a synonym (see also Anderson and Leviton, 1969, p. 51). Stull (1935, p. 407) 
regarded E. j. czarewskii as a synonym of E. niiliaris (Pallas) . 

This species is known from the Kopet Dagh in southern Turkmen, USSR, 
and adjacent northern Iran, 

Eryx johnii (Russell). 

Boa johnii Russell, 1801, Indian Serp., vol. 2, pp. 18 and 20, pis. 16-17, fig. 1 [pi. 17] (type 

locality: Tranquebar, India). 
Eryx johnii, Dumerll and Bibron, 1844, Erp. Gen., vol. 6, p. 458. 



Vol. XXXVIII] LEVITON & ANDERSON: AFGHANISTAN HERPETOFAUNA 195 

Distribution. Known definitely from the coastal plain of West Pakistan 
from the Hab River Valley eastward into the Thar Desert and northward in the 
Indus Valley to central Sind at elevations below 500 feet (Minton, 1966, p. 119). 
According to Smith (1943, p. 114), it occurs also in Rajputana, Punjab, United 
Provinces, Baluchistan, and the Northwest Frontier Provinces (West Paki- 
stan). It must be pointed out that the type of Eryx persicus Nikolsky, 
1907, regarded by Stull (1935, p. 407) and Smith {loc. cit.) as a sub- 
species of E. johnii, is from Aguljaschker, Arabistan [=Khuzestan], Iran. 
There are no further records of the occurrence of either nominal form in the 
intervening areas of Iran. The single record of E. johnii for Afghanistan is that 
of Murray (1892, p. 71) for Mundi, Hissar, south of Kandahar. 

Eryx tataricus (Lichtenstein). 

Boa tatarica Lichtenstein, 1823, in Eversmann, Reise nach Buchara, p. 146 (type locality: 

Aral Sea, USSR). 
Eryx tataricus, Terentjev and Chernov, 1949, Diag. Rept. Amph., p. 230. 

Distribution. Kazakhstan, Uzbekistan, Turkmen, Tajikistan, western 
China, Mongolia, Iran, and Afghanistan; from the Aral Sea east to the Altai 
Mountains, and south to northwestern Baluchistan. It is known from northern 
Afghanistan, Paghman, and the southern desert region of Afghanistan. Boulen- 
ger's records (1889, p. 101) of E. jacidiis from northwestern Afghanistan prob- 
ably should be referred to E. tataricus. 

Family Colubridae 
Genus Boiga Fitzinger 

Boiga Fitzinger, 1826, Neue Class. Rept., pp. 29, 31, 60 (type species: Coluber irregularis 
Merrem, 1820, by original designation). 

Boiga trigonata melanocephalus Annandale. 

Boiga trigonata var. melanocephalus Annandale, 1904, Jour. Asiat. Soc. Bengal, vol. 73, p. 

209, pi. 9, figs. 3^ (type locality: Preso-Baluchistan frontier; 3 syntypes in Indian 

Museum, Calcutta). 

Distribution. From western Baluchistan through eastern Iran, southern 
and western lowland Afghanistan to Uzbekistan and Tajikistan; north in the 
USSR to the Repetek Station, east to the vicinity of Kurgan-Tyube (Tajikistan). 

Genus Coluber Linnaeus 

Coluber Linnaeus, 1758 (in part), Syst. Nat., ed. 10, vol. 1, p. 216 (type species: Coluber 
constrictor Linnaeus, 1758, by subsequent designation by Fitzinger, 1843, Syst. Rept., 
p. 26). 

Coluber karelinii Brandt. 

Coluber karelinii Brandt, 1838, Bull. Acad. Imp. Sci. St. Petersbourg, vol. 3, p. 243 (type 
locality: borders of Caspian Sea; syntypes: Zoological Institute Leningrad nos. 1695- 
1700). 



196 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

Distribution. Eastern Iran, Baluchistan in the Quetta-Pishin area, southern 
lowland Afghanistan, north along the western margin to Turkmen, Uzbekistan, 
Kirghizia, Tajikistan (east to the Vakhsh Valley and Leninabad), southwestern 
Kazakhstan. Kaidak Gulf and southern coast of Aral Sea are northernmost 
known localities (Terentjev and Chernov, 1949, p. 243). 

Coluber ravergieri Menetries. 

Coluber ravergieri Menetries, 1832, Cat. rais. Obj. Zool., p. 69 [not seen] (type locality: 
Baku, Georgia, USSR; holotype in Leningrad). 

Distribution. Extreme northeastern Africa, through the eastern Mediter- 
ranean, including Israel, Lebanon, Syria, Turkey, to Jordan, Iraq, Iran, Georgia, 
Armenia, Azerbaidzhan, Dagestan, USSR, east through Turkmen and Afghani- 
stan to the mountainous areas of the northern part of Kalat District to Chitral, 
West Pakistan. In Soviet Central Asia it reaches as far north as the Emba River 
and the lower reaches of the Syr Darya River; in the northeast it reaches western 
Mongolia. Afghan locahties are in the north and east of the country at elevations 
of 2000-8000 feet. 

Coluber rhodorachis (Jan). 

Zamenis rhodorachis Jan, 1863, vol. 1, p. 356 (type locality: Schiraz, Persia). 
Coluber rhodorachis, Parker, 1931, Ann. Mag. Nat. Hist., ser. 10, vol. 8, p. 516. 

Distribution. Egypt south to SomaUa, east through Arabia, Israel, Jordan, 
Syria, Iraq, Iran, Afghanistan, West Pakistan except Thar Desert and upper 
Indus basin, north to southern Turkmen, southern Uzbekistan, and western 
Tajikistan, USSR, no further north than 40° N. According to Terentjev and 
Chernov (1949, p. 242) the most eastern and northeastern localities are: vicinity 
of Samarkand and Tashkent, Zeravshan Ridge, surroundings of Stalinabad and 
Darvaz Ridge. In Afghanistan it is known from lower elevations on both the 
north and south side of the Hindu Kush. 

*Coluber ventromaculatus Gray and Hardwicke. 

Coluber ventroinaculatus Gray and Hardwicke, 1834, Illust. Indian Zool., vol. 2, pi. 80, fig. 
1 (type locality not stated). 

Distribution. From Almora, United Provinces south to the Khandesh 
District near Bombay, India, west through West Pakistan, Iran, Iraq, northern 
Arabia, and Jordan to Israel. Boulenger (1890, pp. 325-326) includes Afghani- 
stan in the distribution, although we find no documented records. Terentjev 
and Chernov (1949, p. 242) say that records for this species in the USSR refer 
to C. rhodorachis. Various authors have considered both C. karelinii and C. 
rhodorachis as synonyms of C. ventromaculatus, and the separation of these 
forms is far from clear (Leviton, 1959, pp. 454-456). 



Vol. XXXVIII] LEVITON & ANDERSON: AFGHANISTAN HERPETOFAUNA 197 

Genus Eirenis Jan 

Eirenis Jan, 1863, Arch. Zool., vol. 2, p. 256 (type species: Coluber collaris Menetries, by 
subsequent designation by Smith, 1943, Fauna British India, Serp., p. 187). 

*Eirenis persica (Anderson). 

Cyclophis persicus John Anderson, 1872, Proc. Zool. Soc. London, p. 392 (type locality: 

Bushire, Iran; holotype in British Museum). 
Eirenis persica, Stickel, 1951, Herpetologica, vol. 7, p. 128. 

Distribution. The Iranian Plateau, from Jarmo, eastern Iraq, through Iran 
to the Kopet Dagh in southern Turkmen, east to Swat, Punjab, and western Sind, 
West Pakistan. As yet there are no Afghan records; it is included here on the 
basis of its occurrence to the west, south, and east of that country. 

Genus Elaphe Fitzinger 

Elaphe Fitzinger, 1833, in Wagler, J., Descr. Icon. Amphib., vol. 3, text to pi. 27 (type 
species E. parreysii Fitzinger) . 

Elaphe dione (Pallas). 

Coluber dione Pallas, 1773, Reise russ. Reichs, vol. 2, p. 717 (type locaHty: "Salt steppes 

toward the Caspian Sea"). 
Elaphe dione, Dumeril, Bibron, and Dumeril, 1854, Erp. Gen., vol. 7, p. 248. 

Distribution. From the valley of the Volga River, in southeastern Russia 
through temperate Asia to the Amur country in the north and south through 
eastern China to Kuikiang (Stejneger, 1907, p. 318). In Afghanistan known 
from one unpublished record from Bolla-Kuchi village, 6.5 miles southeast of 
Kunduz, Kunduz Province (USNM 166774-166775). 

Genus Lycodon Boie 

Lycodon H. Boie, 1826, Ferussac's Bull. Sci. Nat., vol. 9, p. 238 (type species: Coluber aulicus 
Linnaeus, 1758, by subsequent designation by Fitzinger, 1826, Neue Class. Rept., p. 30). 

*Lycodon striatus bicolor (Nikolsky). 

Contia bicolor Nikolsky, 1903, Ann. Mus. Zool. Acad. Imp. Sci. St. Petersbourg, vol. 8, 
pp. 96-97 (syntypes from eastern Iran, and Kulkulab, Transcaspia, USSR; syntypes: 
Zoological Institute Leningrad nos. 10006, 10013). 

Lycodon striatus bicolor, Chernov, 1935, Compt. Rendu Acad. Sci. URSS, n. ser., vol. 3, 
p. 189. 

Distribution. Eastern and northeastern Iran, southern Turkmen, Uzbeki- 
stan (exclusive of the Kara-Kalpak ASSR), and western Tajikistan; north to 
the Chirchik River hydroelectric plant (Terentjev and Chernov, 1949, p. 238). 
Minton (1966, p. 132) referred a specimen from the vicinity of Quetta, Baluchi- 
stan to this subspecies. Thus far there are no Afghan records for this species; 
it has been taken in the Iranian part of the Seistan Basin, however. 



198 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

Genus Lytorhynchus Peters 

Lytorhynchus Peters, 1862, Monat. Acad. Berlin, p. 273 (type species: Heterodon diadema 
Dumeril, Bibron and Dumeril, 1854, by monotypy). 

Lytorhynchus maynardi Alcock and Finn. 

Lytorhynchus maynardi Alcock and Finn, 1896, Jour. Asiat. Soc. Bengal, vol. 65, p. 562, pi. 
14 (type locality: Koh-Malik-do-Khand, Afghan-Baluchistan frontier; syntypes in the 
Indian Museum, Calcutta, and British Museum). 

Distribution. Desert basins of southern Afghanistan and northern Baluchi- 
stan, West Pakistan (from Nushki to the Iranian border). 

Lytorhynchus ridgewayi Boulenger. 

Lytorhynchus ridgewayi Boulenger, 1887, Ann. Mag. Nat. Hist., ser. 5, vol. 20, p. 413 (type 
locality: Chinkilok, Afghanistan; holotype in British Museum). 

Distribution. From southern and central Iran and Turkmen, USSR (east 
to the Repetek Station) east through northwestern and southern Afghanistan and 
northern Baluchistan. 

Genus Natrix Laurenti 

Natrix Laurenti, 1768, Syn. Rept., p. 73 (type species: Natrix torquatus, by subsequent 
designation of Fleming, 1822). 

Natrix tessellata tessellata (Laurenti). 

Coronella tessellata Laurenti, 1768, Syn. Rept., p. 87 (type locality: "in Japidia, volgo 

Cars"). 
Natrix tessellata, Bonaparte, 1834, Icon. Faun. Ital., vol. 2, pi. 
Natrix tessellata tessellata, Sochurek, 1956, Burgenl. Heimatbl. Eisenstadt, vol. 18, p. 89. 

Distribution. From southern and middle Europe, eastward through the 
Balkans, Southwest Asia, southern USSR (southern Ukraine, Crimea, Transcau- 
casian republics, Turkmen, Uzbekistan, Tajikistan, Khirgiz, and Kazakhstan; 
reaching the lower Usa River along the Volga Valley, and north as far as 53-54° 
N.) to northern and eastern Afghanistan and Chitral, West Pakistan. 

Genus Oligodon Boie 

Oligodon Boie, 1827, Isis von Oken, p. 519 (type species: Coluber bitorquatus Reinwardt, 
by monotypy) . 

* Oligodon taeniolatus (Jerdon). 

Coronella taeniolata Jerdon, 1853, Jour. Asiat. Soc. Bengal, vol. 22, p. 528 (type locality: 

Vizagapatam; based on Russell, 1796, Indian Serp., vol. 1, p. 24, pi. 19). 
Oligodon taeniolatus, Wall, 1921, Sn. Ceylon, p. 239. 

Distribution. From Bihar, India, to southeastern Baluchistan, south 
through peninsular India to Ceylon. It is found in West Pakistan in the low- 
lands, from the Indus delta north to Rawalpindi, west to Bela (Minton, 1966, 



Vol. XXXVIin LEVITOX & ANDERSON: AFGHANISTAN HERPETOFAUNA 199 

p. 134). It is known in southern Turkmen from a single specimen from the 
Kopet Dagh (Terentjev and Chernov, 1949, p. 258). No records exist for 
Afghanistan, but its known distribution suggests that it occurs there in the low 
deserts of the south and west. 

Genus Psammophis Fitzinger 

Psammophis Fitzinger, 1826, Neue Class. Rept., pp. 29-30 (type species: Coluber sibilans 
Linnaeus, 1758, by original designation). 

* Psammophis leithii Giinther. 

Psammophis leithii Gunther, 1769, Proc. Zool. Soc. London, p. 505, pi. 39 (type locality: 
Sind; holotype in British Museum). 

Distribution. From Fyzabad, in the United Provinces to Poona and west 
to Waziristan and southeastern Baluchistan. In West Pakistan it is known from 
Azad Kashmir to the southern Thar and west to Waziristan and southern Kalat 
District (Minton, 1966, p. 141). No verifiable records exist for Afghanistan; 
Boulenger's (1889, p. 103) records for Helmand and Hamun to Khusan refer to 
Psammophis schokari (Boulenger, 1896, pp. 157-158). 

Psammophis lineolatus (Brandt). 

Coluber (Taphrometopon) lineolatus Brandt, 1838, Bull. Acad. Imp. Sci. St. Petersbourg, 

vol. 3, p. 243 (type locality: Transcaspia). 
Psammophis lineolatus, Smith, 1943, Fauna British India, vol. 3, p. 367. 

Distribution. Turkmen, Uzbekistan, Tajikistan, Khirgiz, Kazakhstan, 
USSR, to about 49° N., through Mongolia to Kansu and Ala Shan inclusive; 
south through eastern Iran (west as far as Yezd-i-Khast) and Afghanistan to 
the Quetta Plateau, Baluchistan, West Pakistan. 

Psammophis schokari (Forskal). 

Coluber schokari Forskal, 1775, Descr. Anim., p. 14 (type locality: Yemen). 
Psammophis schokari, Boulenger, 1896, Cat. sn. British Mus., vol. 3, pp. 157-158. 

Distribution. From Morocco across North Africa (south to Somalia) and 
the entire Arabian peninsula (at least the coastal regions) through Israel, Leb- 
anon, Syria, Jordan, Iraq, Iran, southern Turkmen, southern Afghanistan, and 
West Pakistan, east to Kashmir, and south to Kutch. 

Genus Ptyas Fitzinger 

Ptyas FiTZLNGER, 1843, Syst. Rept., p. 26 (type species: Coluber blumenbachii Merrem, 1820, 
by original designation). 

Ptyas mucosus (Linnaeus). 

Coluber mucosus Linnaeus, 1758, Syst. Nat., vol. 1, p. 226 (type locality: India; type in 

Stockholm). 
Ptyas mucosus, Cope, 1860, Proc. .^cad. Philadelphia, p. 563. 



200 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

Distribution. From the Murgab basin in southern Turkmen, probably 
locally in eastern Iran, southern Afghanistan (existing records are all in the 
southeastern portion, including the Kabul River Valley, and between Kabul and 
Kandahar), West Pakistan, from the lower Indus west to Baluchistan and north 
to Chitral, throughout India and Ceylon and eastward to southern China and 
Viet Nam; Formosa; Java, Sumatra, and the Andaman Islands. 

Genus Spalerosophis Jan 

Spalerosophis Jan, 1865, in De Filippi, Viag. in Persia, vol. 1, p. 356 (type species: Spaleroso- 
phis microlepis Jan 1865, by monotypy). 

Spalerosophis diadema schirazana (Jan). 

Periops parallelus var. schirazana Jan, 1865, in De Filippi, Viag. in Persia, vol. 1, p. 356 

(type locality: Persia). 
Spalerosophis diadema schirazianus, Mertens, 1956, Jh. Ver. Naturk. WiJrttemberg, vol. Ill, 

p. 96. 

Distribution. From the Zagros Mountains of western Iran east through 
southern Turkmen, Uzbekistan, and western Tajikistan, USSR (extreme locality 
records being: Vakhsh River; vicinity of Osh; coastline of Kenderli Gulf; and 
Kyzyl Kum, north of Khoresm oasis [Terentjev and Chernov, 1949, p. 245]), 
northern lowland Afghanistan, the southern Afghan desert region, north at least 
to Paghman, and south to Quetta and western Las Bela in Baluchistan. 

Xenochrophis piscator: See Addendum, pg. 206. 

Family Elapidae 
Genus Naja Laurenti 

Naja Laurenti, 1768, Synops. Rept., p. 90 (type species: Naja lutescens Laurenti, 1768, by 
subsequent designation by Leviton, 1967, in Buciierl, W., Venomous Animals and their 
Venoms, p. 547). 

Naja oxiana (Eichwald). 

Tomyris oxiana Eichwald, 1831, Zool. Spec, p. 171 (type locality: Transcaspia, USSR). 
Naja oxiana, Strauch, 1869, Bull. Acad. Imp. Sci. St. Petersbourg, vol. 13, cols. 81-94. 

Distribution. Southern Turkmen, Uzbekistan (north to Samarkand and 
Aristan-Bel-tau mountains), southwestern Tajikistan, USSR; northeastern Iran; 
Afghanistan (documented records are in the northwest, south, and south of the 
Hindu Kush) ; northeastern Baluchistan, Northwest Frontier Provinces, and 
Kashmir, West Pakistan. 

Family Leptotyphlopidae 
Genus Leptotyphlops Fitzinger 

Leptotyphlops Fitzinger, 1843, Syst. Rept., p. 24 (type species: Typhlops nit^ricans Schlegel, 
1844, by original designation). 



Vol. XXXVIII] LEVITON & ANDERSON: AFGHANISTAN HERPETOFAUNA 201 

Leptotyphlops blandfordi (Boulenger). 

Glauconia blanfordii Boulenger, 1890, Fauna British India, p. 243 (type locality: Sind; 

holotype in British Museum). 
Leptotyphlops blanjordi, Werner, 1936, Festschr. Strand, vol. 2, p. 201. 

Distribution. Known from the Indus delta north through southern Punjab 
to Sibi and the Northwest Frontier Provinces (Minton, 1966, p. 117). In Afghani- 
stan it occurs in the valley of the Kabul River. Specimens recorded from Kuh-i- 
Malik Siah, at 5000 feet elevation at the point where the Iranian, West Pakistan, 
and Afghan borders meet (Alcock and Finn, 1896, p. 561) have not been reex- 
amined by subsequent workers. 

Family Typhlopidae 
Genus Typhlops Oppel 

Typhlops Oppel, 1811, Ordn. Fam. Gatt. Rept., p. 54 (type species: Angiiis lumbricalis 
Linnaeus, 1766 [jide Smith, 1943, p. 43]). 

Typhlops vermicularis Merrem. 

Typhlops vermicularis Merrem, 1820, Tent. Syst. Amph., p. 158 (type locality: Greek 
islands; restricted by Mertens and Muller, 1928, Abh. Senckenb. Naturf. Ges., vol. 41, 
p. 45). 

Distribution. Southern Yugoslavia, Albania, southern Bulgaria, Greece, 
Noxos in the Cyclades, Rhodes, Turkey, Syria, lower Egypt, Israel, Transcau- 
casian republics of the USSR, Dagestan, southern Turkmen, southern Uzbekistan, 
.southern and western Tajikistan, Iran, and Afghanistan (the only known record 
being for north of Herat, in the northwest). 

Family \"iperidae 

Genus Agkistrodon Beauvois 

Agkistrodon Beauvois, 1799, Trans. American Philos. Soc, vol. 4, p. 381 (type species: 
Agkistrodon mokasen Beauvois, 1799, by monotypy). 

Agkistrodon halys (Pallas). 

Coluber halys Pallas, 1776, Reise versch. Prov. Russ. Reich, vol. 3, p. 403 (type locality: 

eastern Naryn Steppe). 
Ancistrodon halys, Boulenger, 1896, Cat. sn. British Mus., vol. 3, pp. 524-525. 

Distribution. From Azerbaidzhan, northern Iran, and the Trans-Volga 
region through central Asia and southern Siberia (to about 55-56° N.), to the 
Pacific. Our inclusion of this species in the Afghan fauna is based on a specimen 
in the Universitetets Zoologiske Museum, Copenhagen (not seen by us) labeled 
"Ancistrodon" from the Sauzak Pass near Herat, northwestern Afghanistan. 

Agkistrodon himalayanus (Giinther). 

Halys himalayanus Gunther, 1864, Rept. British India, p. 393, pi. 24, fig. A (type locality: 

Garwal, western Himalayas; 2 syntypes in British Museum). 
Ancistrodon himalayanus, Boulenger, 1890, Fauna British India, p. 424, fig. 125. 



202 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

Distribution. The western Himalayas, from Sikkim to Chitral, West Paki- 
stan. A specimen labeled "Ancistrodon" (not seen by us) in the Universitetets 
Zoologiske Museum, Copenhagen, collected at Wama, Nuristan, in eastern Af- 
ghanistan, prompt our inclusion of this species in the checklist. 

Genus Echis Merrem 

Echis Merrem, 1820, Tent. Syst. Amph., p. 149 (type species: Pseudoboa carinata Schneider, 
1801, by subsequent designation by Fitzinger, 1843, Syst. Rept., p. 28). 

Echis carinatus (Schneider). 

Pseudoboa carinata Schneider, 1801, Hist. Amph., vol. 2, p. 285 (type locality: Arni; based 

on Russell, 1796, Indian Serp., vol. 1, pi. 2). 
Echis carinata, Wagler, 1830, Syst. Amph., p. 177. 

Distribution. Northern Africa south to Ghana in the west, Kenya in the 
east, through Arabia and Southwest Asia north to Turkmen, southern Uzbekistan 
(north to Samarkand), and northwestern Tajikistan, east to the Vakhsh Valley 
inclusive in the USSR, south through Afghanistan and all of West Pakistan ex- 
clusive of the Himalayan region, east to the Ganges Valley, and south through 
drier regions of India, to northern Ceylon. In Afghanistan it occurs in the lower 
elevations both north and south of the Hindu Kush. 

Genus Eristicophis Alcock and Finn 

Eristicophis Alcock and Finn, 1896, Jour. Asiatic Soc. Bengal, vol. 65, p. 564 (type species: 
Eristicophis macmahoni Alcock and Finn, by monotypy). 

Eristicophis macmahoni Alcock and Finn. 

Eristicophis macmahoni Alcock and Finn, 1896, Jour. Asiatic Soc. Bengal, vol. 65, pp. 564- 
565, pi. 15, figs. 1, la (types from Amirchah ; Zeh ; Drana Koh ; Robat I, Afghan- 
Baluch border). 

Distribution. The desert basin region of northwestern Baluchistan, West 
Pakistan, from Nushki westward to the border, and south to Kharan; the south- 
ern desert of Afghanistan in Registan and the Dasht-i-Margo; Seistan, eastern 
Iran; below 4000 feet elevation. 

Genus Pseudocerastes Boulenger 

Pseudocerastes Boulenger, 1896, Cat. sn. British Mus., vol. 3, p. 501 (type species: Cerastes 
persicus Dumeril, Bibron and Dumeril, 1854, by monotypy). 

*Pseudocerastes persicus persicus (Dumeril, Bibron, and Dumeril). 

Cerastes persicus Dumeril (AHA), 1853, Mem. Acad. Sci. Inst. France, vol. 23, p. 532 
(nomen nudum) ; Dumeril, Bibron, and Dumeril, 1854, Erp. Gen., vol. 7, p. 1443, pi. 
78b, fig. 5 (type locality: Persia). 

Pseudocerastes persicus, Boulenger, 1896, Cat. sn. British Mus., vol. 3, p. 501. 

Distribution. Pseudocerastes persicus ranges from the central Sinai penin- 
sula through the Negev in Israel, Jordan, northern Saudi Arabia, Iraq, Iran, and 



Vol. XXXVIII] LEVITON & ANDERSON: AFGHANISTAN HERPETOFAUNA 203 

West Pakistan as far east as Manguli in southwestern Kalat. There is a frag- 
mentary specimen {P. bicornis Wall) from Khajieri Kach above Gwaleri Kolal in 
the Gomal Pass, Waziristan, West Pakistan. This latter record is close to the 
Afghan border, but while Afghanistan has been included in the distribution of 
this species by previous authors, we find no documented records. 

Genus Vipera Laurenti 

Vipera Laurenti, 1768, Synops. Rept., p. 99 (type species: Vipera redi, Latreille, by sub- 
sequent designation by Fitzinger, 1843, Syst. Rept., p. 28). 

Vipera lebetina (Linnaeus). 

Coluber lebetinus Linnaeus, 1758, Syst. Nat., ed. 10, vol. 1, p. 216 (type locality: Cyprus; 
restricted by Mertens and Muller, 1928, Abh. Senckenberg. Naturf. Ges., vol. 41, p. 52). 
Vipera lebetina, Daudin, 1803, Hist. nat. Rept., vol. 6, p. 137. 

Distribution. North Africa from Morocco to TripoU; Cyprus and the 
Cyclades Islands of the eastern Mediterranean, Turkey, Syria, Lebanon, Israel, 
Jordan, Iraq, Iran, Transcaucasian republics of the USSR, Dagestan, southern 
part of Central Asian repubhcs, east to Khorog and northeast to the vicinity of 
Leninabad, USSR; Afghanistan, West Pakistan (from Waziristan south to the 
Quetta Plateau), and east to Kashmir. Apart from a recently collected specimen 
from Jalalabad, in the valley of the Kabul River, eastern Afghanistan, all Afghan 
records for this species are from the northwestern part of the country. 

BIBLIOGRAPHY 

Note: The following list includes not only those titles cited in the text but 
also those principally concerned with the Afghanistan herpetofauna and which 
serve as a primary source of original locality data. Primary systematic sources 
include, in addition to those cited below, the references noted in the synonymies 
in the check list. 

Alcock, a., and F. Finn 

1897. An account of the Reptilia collected by Dr. F. P. Maynard, Captain A. H. 
McMahon, CLE., and the members of the Afghan-Baluch Boundary Com- 
mission of 1896. Journal of the Asiatic Society of Bengal, vol. 65, pp. 550-566, 
pis. 11-15. 
Anderson, Steven C, and Alan E. Leviton 

1967a. A new species of Eremias (Sauria: Lacertidae) from Afghanistan. Occasional 
Papers of the California Academy of Sciences, no. 64, 4 pp. 

1967b. A new species of Phrynocephalus (Sauria: Agamidae) from Afghanistan, with 
remarks on Phrynocephalus ornatus Boulenger. Proceedings of the California 
Academy of Sciences, ser. 4, vol. 35, pp. 227-234. 

1969. Amphibians and reptiles collected by the Street Expedition to Afghanistan, 1965. 
Proceedings of the Cahfornia Academy of Sciences, ser. 4, vol. 37, pp. 25-56. 



204 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

BOULENGER, GeORGE A. 

1889. Reptiles and batrachians. In: Aitchison, J. E. T., The zoology of the Afghan 

Delimitation Commission. Transactions of the Linnean Society of London, ser. 
2, vol. 5, Zoology, pp. 94-106, pis. 8-11. 

1890. Reptilia and Batrachia. In: The Fauna of British India. . . London, xviii + 541 

pp. 

1896. Catalogue of the snakes in the British Museum (Natural History). Volume 3. 

London, xiv + 727 pp., 25 pis. 
Bruck, Gustav 

1968. Zur herpetofauna Afghanistans. Vestnik Ceskoslovenske Spolecnosti Zoologicke 

Acta Societatis Zoologicae Bohemoslovacae, vol. 32, pp. 201-208. 
Clark, Richard J., Erica D. Clark, Steven C. Anderson, and Alan E. Leviton 

1969. Report on a collection of amphibians and reptiles from Afghanistan. Proceedings 

of the California Academy of Sciences, ser. 4, vol. 36, pp. 279-316. 

FiTZINGER, LeOPOLDO 

1843. Systema reptilium. Vindobonae, 106 + vi+r3] pp. 
Haas, Georg 

1957. Some amphibians and reptiles from Arabia. Proceedings of the California Acad- 
emy of Sciences, ser. 4, vol. 29, pp. 47-86. 
Kullmann, E. 

1967. Wozu bauen wir einen Zoo in Kabul. Freunde des Kolner Zoo, Jahrg. 10, 
pp. 43-49. 
Lantz, L. a. 

1918. Reptiles from the River Tajan (Transcaspia). Proceedings of the Zoological 
Society of London, 1918, pp. 11-17, pi. 1. 
Leviton, Alan E. 

1959. Report on a collection of reptiles from Afghanistan. Proceedings of the California 
Academy of Sciences, ser. 4, vol. 29, pp. 445-463. 
Leviton, Alan E., and Steven C. Anderson 

1961. Further remarks on the amphibians and reptiles of Afghanistan. Wasmann Journal 

of Biology, vol. 19, pp. 269-2 76. 
1963. Third contribution to the herpetology of Afghanistan. Proceedings of the 
California Academy of Sciences, ser. 4, vol. 31, pp. 329-339. 
Mertens, Robert 

1965. Bemerkungen ijber einige Eidechsen aus Afghanistan. Senckenbergiana Biologia, 

vol. 46, pp. 1^. 
MiNTON, Sherman A. 

1966. A contribution to the herpetology of West Pakistan. Bulletin of the American 

Museum of Natural History, vol. 234, pp. 29-184, pis. 9-36. 
Murray, J. A. 

1892. The zoology of Beloochistan and southern Afghanistan. London, 83 pp. 
NiKOLSKY, Alexander M. 

1897. Les reptiles, amphibiens, et poissons recueillis par M. N. Zarudny dans la Perse 

orientale. Annuaire du Musee Zoologique de I'Academie Imperiale des Sciences 
de St. Petersbourg, vol. 2, pp. 306-348. 
1899. Reptiles, amphibiens, et poissons recueillis pendant le voyage de M. N. A. 
Zarudny en 1898 dans la Perse. Annuaire du Musee Zoologique de I'Academie 
Imperiale des Sciences de St. Petersbourg, vol. 4, pp. 375-415. 



Vol. XXXVIII] LEVITON & ANDERSON: AFGHANISTAN HERPETOFAUNA 205 

Smith, Malcolm A. 

1935. Sauria. In: The Fauna of British India. . . Reptilia and Amphibia. Volume 2. 

London, xiii + 440 pp., 1 pi., 1 map. 
1940. Contributions to the herpetology of Afghanistan. Annals and Magazine of 

Natural History, ser. 11, vol. 5, pp. 382-384. 
1943. Serpentes. In: The Fauna of British India. . . Reptilia and Amphibia. Volume 
3. London, xii + 583 pp., 1 map. 
Stejneger, Leonhard 

1907. Herpetology of Japan and adjacent territory. Bulletin of the U. S. National 
Museum, no. 58, xx + 577 pp., 35 pis. 
Stull, Olive 

1935. A checklist of the family Boidae. Proceedings of the Boston Society of Natural 
History, vol. 40, pp. 387-408. 
Taylor, Edward H. 

1935. A taxonomic study of the cosmopolitan scincoid lizards of the genus Eiimeces. . . 
Bulletin of the University of Kansas, vol. 36, 643 pp., 43 pis. 
Terentjev, Paul V., and Sergius A. Chernov 

1949. Key to amphibians and reptiles. Moscow, ed. 3, 339 pp. (Translated into English 
in 1965; U. S. Department of Commerce, Washington, D. C. 315 pp. Original 
pagination indicated in text.) 
Ueno, Shun-Ichi, and Kenji Nakamura 

1966. The anurans collected by the Kyoto University Pamir-Hindukush Expedition 
1960. Results of the Kyoto University Scientific E.xpedition to the Karakoram 
and Hindukush, 1955, vol. 8, p. 327. 
Wermuth, Heinz 

1965. Liste der rezenten .'\mphibien und Reptilien. Gekkonidae, Pygopodidae, Xantusi- 
idae. Das Tierreich, Lief. 80, xxii + 246 pp. 
Wettstein-Westersheimb, Otto von 

1960a. Contribution a I'etude de la faune d' Afghanistan. 3. Lacertilia aus .\fghanistan. 

Zoologischer Anzeiger, vol. 165, pp. 58-63. 
1960b. Drei seltene Echsen aus Siidwest-Asien. Zoologischer Anzeiger, vol. 165, pp. 
190-193. 



ADDENDUM 
*Agamura femoralis Smith. 

Agamura jemoralis Smith, 1933, Rec. Indian Mus., vol. 35, p. 17 (type locality: Kharan, Baluchistan, West 
Pakistan). 
Distribution. Northwestern Baluchistan, West Pakistan. 

Genus Alsophylax Fitzinger 

Alsophylax Fitzinger, 1843, Syst. Rept., p. 18 (type species Gymnodnctylus pipiens Eichwald, by original 
designation). 

Alsophylax cf. pipiens (Pallas). 

Lacerta pipiens Pallas, 1811, Zoogr. Ross-asiatica, p. 27 (type locality: Mt. Bogdo, near the Volga River, 

USSR). 
Alsophylax pipiens Fitzinger, 1843, Syst. Rept., pp. 18, 90. 

Distribution. From the lower Volga and Transcaspia region of the USSR to central 
MongoHa and the Ala-Schan mountains, south to northern Iran and Afghanistan. In 
Afghanistan we have seen material only from the vicinity of Kabul (see: Clark, Clark, 



206 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

Anderson and Leviton, 1969). Mertens (1965, p. 2) reports one specimen from Oukak. 

Genus Xenochroj>his Gimther 

Xcnochrophis Gunther, 1864, Rept. British India, p. 273 (type species: Psammophis cerasogaster Cantor, 
by monotypy). 

Xenochrophis piscator (Schneider). 

Hydnis piscator Schneider, 1799, Hist. Amphib., p. 247 (type locality: "Indiae orientalis," based on 

Russell's "Neeli Koea"). 
Tropidonotus piscator Boulenger, 1890, Fauna British India, Rept. & Batr., p. 349 (in part). 
Natrix piscator Smith, 1943, Fauna British India, Serp., p. 293. 
Xenochrophis piscator Maln..\te, 1965, Proc. Acad. Nat. Sci., Philadelphia, vol. 117, p. 19. 

Distribution. From Baluchistan, West Pakistan, throughout all of India, Ceylon, 
central Nepal to 5000 feet, and east throughout the whole of the Indochinese Subregion, 
Southern China, Malaya and western Indonesia. In Afghanistan, collected 40 km. south- 
west of Jalalabad. 



NOTE. In a recent paper entitled "Notes on the herpetofauna of certain provinces of 
Afghanistan," (Zoologiske Listy, vol. IS, pp. 55-66, 1969) Dr. B. Krai documents the 
occurrence of Psammophis leithi in Afghanistan. His specimen came from 8 km. from 
Jalalabad, toward Sarsahi. He also adds two hitherto unrecorded species to the faunal list 
of the country: OUgodon arnensis (Family Colubridae), and Bnngarus caeruleus (Family 
Elapidae). 



PROCEEDINGS 

OF THE 

CALIFORNIA ACADEMY OF SCIENCES 

FOURTH SERIES 



Festschrift for George Sprague Myers 



Vol. XXXVIII, No. 11, pp. 207-214; 1 fig.; 1 map. December 31, 1970 



EXPLOSIVE SPREAD OF THE ORIENTAL 

GOBY ACANTHOGOBIUS FLAVIMANUS 

IN THE SAN FRANCISCO BAY-DELTA 

REGION OF CALIFORNIA 

By 

Martin R. Brittan 

Department of Biological Sciences, Sacramento State College, Sacramento, California 

and 

John D. Hopkirk 
Department of Biological Sciences, Sonoma State College, Cotati, California 

and 

Jerrold D. Conners 
School of Public Health, University of California, Berkeley, California 

and 

Michael Martin 
Department of Zoology, Unix'ersity of Southern California, Los Angeles, California 

One of the features of animal distribution in recent years is the unexpected 
appearance of a species from far-off lands which rapidly, sometimes explosively, 
expands its range in its new homeland, often in direct competition with estab- 
hshed elements of the native fauna. 

[207] 



208 



CALIFORNIA ACADEMY OF SCIENCES 



[Proc. 4th Ser. 




Map 1. Collection localities for Acanthogobhis flavimanus in the San Francisco Bay and 
Sacramento-San Joaquin Delta regions, 196.5-1968 (the many verbal reports by fishermen in 
1968 and later are not recorded). 1. Alviso, Santa Clara County. 2. Palo Alto Yacht Harbor, 
Santa Clara County. 3. Foster City Lagoon, San Mateo County. 4. San Mateo Bridge, San 
Mateo-Alameda counties. S. Plummer Creek, near Newark, Alameda County. 6. Lake Mer- 
rit, Oakland, Alameda County. 7. Aquatic Park, Berkeley, Alameda County. 8. Treasure 
Island, San Francisco-Alameda counties; Angel Island, Marin County. 9. Tiburon, Paradise 
Cay, Belvedere, Marin County. 10. Richardson Bay, Belvedere, Marin County. 11. Marin 
Islands, San Rafael, Lower San Pablo Bay, Marin County. 12. Bolinas Lagoon, Solano 
County. 13. Napa Slough, Solano County. 14. Benicia, Carquinez Strait, Solano County. 
IS. Suisun Bay, Solano County. 16-19. Montezuma Slough, Solano County. 20. Suisun 
Bay, Solano County. 21. Antioch, San Joaquin River, Contra Costa County. 22. Snodgrass 
Slough off Sacramento River, opposite Walnut Grove, Sacramento County. 23. Prisoners 
Point, San Joaquin County. 24. Stockton (Deep Water Channel), San Joaquin County. 
25. Tracy Pumping Plant, Alameda County. 26. Delta-Mendota Canal at Newman Waste- 
way, Stanislaus County. 27. San Luis Reservoir, Merced County. 



In 1963 two specimens of Acanthogobiits flavimanus Teniminck and Schlegel, 
a euryhaline goby of Japan and adjacent mainland w^aters, were taken in 
the Sacramento-San Joaquin River Delta of California. The first specimen of 
the "mahaze" (its Japanese name; Okada, 1960) was taken on January 18 at 
Prisoners Point on Venice Island, and the second was taken on March 29 in the 
Stockton Deepwater Channel at the entrance of the Calaveras River, just below 



Vol. XXXVIII] BRITTAX ET AL.: ACANTHOGOBIUS FLAVIMANUS 



209 






:>^. 




Figure 1. Acanthogobius flavimamis (Temminck and Schlegel). 123 mm. standard 
length, from the San Joaquin River at Prisoners Point, Venice Island, San Joaquin County 
(from Brittan, Albrecht, and Hopkirk, 1963). The higher outline of the dorsal fin added to 
the drawing is from a specimen of 176 mm. standard length from the Delta-Mendota Canal 
at Newman Wasteway, Stanislaus County, and illustrates the condition typical in larger 
individuals. 



the Port (and city) of Stockton. The first 2 specimens were 123 and 69 mm. 
standard length, respectively. 

No further examples were collected until late 1964 when several were taken 
from Palo Alto Yacht Harbor (Robert Hassur, verbal communication) and from 
a trap in Leslie Salt Company evaporation ponds at Alviso (4; 141-153 mm.). 
These localities are roughly 80-90 water miles from the initial collection points. 

No further specimens were taken until August, 1965, when one was obtained 
off Marin Island near San Rafael, INIarin County (male, 161 mm.) and, surpris- 
ingly enough at the time, an additional one at Newman wasteway on the Delta- 
Mendota Canal, which carries fresh water for irrigation from the Delta to the 
central San Joaquin Valley. The latter locality is approximately 80 water miles 
from the other farthest point from which the mahaze had been collected : Alviso, 
Santa Clara County. 

In 1966, specimens of A. jlavimanus were taken from widely spread localities 
and in increasing numbers, a trend which persisted during 1967. Starting in 
May, 1966, Jerrold Conners, using a small trawl mainly around Treasure Island, 
which lies adjacent to Yerba Buena Island between San Francisco and Oakland, 
took a total of 65 gobies on 12 different collecting trips; the largest number 
taken at any one time was 27. Specimens were also taken in San Francisco Bay 
in 1966 in the San Rafael Channel in May (1 male; 149 mm.), in the lower tidal 
reaches of Plummer Creek near Newark in May (3; 24-31 mm.), July (2; 44- 
64 mm.), and August (9; 80-108 mm.), and off Treasure Island in September 
(4 males and 2 females: 102-125 mm.), October (7 males and 1 female; 132-176 
mm.), and November (1 male and 2 females; 124-129 mm., and 1 unsexed speci- 
men of 115 mm.). A specimen was also taken in Richardson Bay in October 
(119 mm.) . The small size of the first Plummer Creek specimens indicates breed- 



210 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

ing at that locality. J. A. Aplin of the California Department of Fish and Game 
(personal communication) notes that the Department's research vessel Nauti- 
lus, taking monthly samples at 6 stations in San Francisco Bay from the San 
Rafael Bridge southward 20 nautical miles to Dunbarton Bridge collected no 
Japanese gobies during the first 3 years of a biological survey of the Bay begin- 
ning in 1963, but during the fourth year (1966) took 10 specimens near Angel 
Island in September, 3 just south of the San Mateo Bridge ( 1 each in August, 
October, and November), and 1 just south of Dunbarton Bridge in November. 
Other material was collected in San Pablo Bay in May at McNear Beach (3 
males; 139-168 mm.) and in November at Napa Slough (2 females and 1 male; 
92-123 mm.), in Carquinez Strait between San Pablo and Suisun Bay at Beni- 
cia (4 males and 1 female; 107-177 mm.), and in the Delta at Antioch (1 male; 
114 mm.). In March, a single adult was taken from the screen of the Tracy 
Pumping Plant; young fish would easily pass through the screen as water is 
pumped into the Delta-Mendota Canal. 

During 1967 more records poured in. Fifty-five fish were taken in January 
between Angel Island and Treasure Island, and exhibited alive for several months 
in pure seawater at 50°F. at Steinhart Aquarium in San Francisco; these speci- 
mens were fully adult. In February, 7 females (117-155 mm.) were taken off 
Treasure Island and 20 females and 1 male ( 1 13-157 mm.) were collected east of 
San Rafael. In September, 5 were taken at Foster City Lagoon near the west end 
of San Mateo Bridge, and in December, 14 were taken at several locations in 
Suisun Bay and adjacent Montezuma Slough by the CaHfornia Department of 
Fish and Game. During 1967, specimens also were taken in Lake Merritt, a tidal 
lake in Oakland, in Belvedere Lagoon near Belvedere-Tiburon, at Aquatic Park 
in Berkeley, at Paradise Cay on the Tiburon Peninsula, and at other localities 
on San Francisco Bay. 

Two surprises came to hght in 1967. In July, checking operations by the 
CaUfornia Department of Fish and Game noted approximately 10,000 dead 
"trash" fishes in the San Luis Reservoir in Merced County, as a result of total 
depletion of oxygen because of an algal bloom and following die-off. This is a 
large, recently filled, man-made reservoir behind a gigantic earth-fill dam, and is 
located in the arid foothills of the inner Coast Range about 100 air miles south- 
east of San Francisco. It receives fresh water from the Delta (and eventually 
from the Feather River, a tributary of the Sacramento) via the CaUfornia Aque- 
duct of the California Water Project and the Delta-Mendota Canal of the Central 
Valley Project. About half of the kill consisted of .1. Jlavimanusl The balance 
were bluegills (Lepoinis macrochirus), crappie (Pomoxis), and sticklebacks 
{Gasterosteus aculeatus). In December, a single goby (215 mm. total length) 
was taken from lower (tidal) Pine Gulch Creek in Bolinas Lagoon. This lagoon 
has no connection with San Francisco Bay except by approximately 15 miles of 
open rocky seacoast. 



Vol. XXXVIII] BRITTAX ET AL.: ACAXTHOGOBIUS FLAVIMANUS 211 

During 1968 the mahaze continued to be taken by biologists and fishermen 
in San Francisco Bay, in San Pablo and Suisun bays, and the Delta. Two collec- 
tions indicate the species is spreading northward from the Delta in fresh water. 
In August, 1968, a specimen 95 mm. long was taken from Snodgrass Slough, a 
tributary of the Sacramento River near Walnut Grove (the mahaze is said by 
fishermen to be "common" here; Robert McKechnie, California Department of 
Fish and Game, personal communication). In October, an example 100 mm. in 
length was collected in the Sacramento Ship Channel just south of the Port and 
City of Sacramento. 

Most of the collections made during 1963-67, which delineated the buildup 
of Acanthogobius jlavimanus from 1 specimen or a few specimens taken at widely 
scattered localities to specimens taken nearly everywhere in the bays and the 
Delta, frequently in considerable numbers, were made by Conners while trolling 
for English sole in San Francisco Bay, by Al Aplin during the California Depart- 
ment of Fish and Game's biological survey of San Francisco Bay, and by the 
Department's Delta Study team. These collections and others, between 1963 
and 1967, indicate a slow buildup period in which the goby was steadily increas- 
ing its numbers while wandering greatly (specimens taken from widely separated 
areas with many young of the year and of the previous year caught), followed by 
a great increase after widespread establishment. During 1966 the species appar- 
ently reached nearly the full extent of its distribution in San Francisco Bay and 
the Delta. When the species first gained access to the area it is impossible to 
say, but the first specimen collected in January 1963, was, from its size, probably 
entering its second year of life; the second specimen was a large subadult. It is 
probable they were spawned in the Delta. However, the fact that the species 
was not previously collected in spite of the considerable sport fishery in the 
region (as well as scientific collecting) indicates that the date of introduction 
was probably not more than 3 or 4 years previous to 1963. 

In the central portion of San Francisco Bay the principal goby species taken 
with the mahaze is the bay goby, Lepidogobius Icpidus (Girard), while in the 
collections made in the tidal portion of Plummer Creek, near its exit into south 
San Francisco Bay, it occurred with the mudsucker goby, Gillichthys mirabilis 
Cooper; the arrow goby, Clcvelandia ios (Jordan and Gilbert); and the cheek- 
spot goby, Ilypnus gilberti (Eigenmann and Eigenmann), the last being the most 
common. Salinities in Plummer Creek ranged from 16.9 percent on ]May 26 to 
30.8 percent on August 17. At Palo Alto Yacht Harbor specimens belonging to 
Acanthogobius jlavimanus have been taken since 1964; this species now heavily 
outnumbers the staghorn sculpin, Leptocottus armatus, formerly the commonest 
bottom fish (Robert Hassur, verbal communication). 

The specimen taken one mile west of Antioch Bridge in the San Joaquin 
River was found with the cyprinids Lavinia exilicauda (Baird and Girard), 
Orthodon microlepidotus (Ayres), and Pogonichthys macrolcpidotus (Ayres), the 



212 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

catostomid Catostomus occidentalis Ayres, the embiotocid Hysterocarpus traski 
Gibbons, and small striped bass, Roccus saxatilis (Walbaum) (Serranidae) ; all 
except the last are California lowland freshwater endemics. They were collected 
over a fine sand bottom; the water was fresh, but muddy, with an incoming tide. 
The 16 specimens taken from the Newman wasteway of the Delta-Mendota 
Canal were associated with white catfish, Ictalurus catus (Linnaeus) (1309 
specimens), and with American and threadfin shad, Alosa sapidissima (Wilson) 
and Dorosoma petenensis (Giinther) (1317 specimens, about 4:1 in favor of 
former) ; striped bass, Roccus (350) ; channel catfish, Ictalurus punctatus (Rafi- 
nesque) (7); spHttail, Pogonichthys (2); and tule perch, Hysterocarpus (1). 
The specimens from the Delta-Mendota Canal were taken in freshwater in No- 
vember, 1966, and December, 1965, and were examined by Michael Martin. On 
the basis of scale annulae and standard lengths, they were assigned to age classes: 
class ranged up to 132 mm. for males and 130 mm. for females, class I from 
107-171 mm. for males and 121-162 mm. for females, while 1 male of 177 mm. 
was assigned to class II. The ovaries of 7 class I females were measured by water 
displacement, the ovarian volume increasing from an average of 0.0061 cc. per 
millimeter (of the standard length) in early November (Napa Slough) to 0.015 
cc./mm. in December (Delta-Mendota Canal), though average water tempera- 
ture decreased 5 degrees F. in the interim. This rapid gonadal development 
indicates a spawning season from January to March, similar to that reported for 
Japan (Dotu and Mito, 1955). Sexual maturity in the California population may 
not arrive until the end of the second or third year of hfe. Ripe adults were 
generally scarce in the San Francisco Bay areas sampled by Conners from spring 
1966 to winter of 1966-67. In addition, females predominate in the mahaze catch 
in February, 1967, the males apparently being in shallow water involved in the 
construction of territories and breeding burrows (Dotu and Mito, 1955). Males 
have darker and longer median fins, but do not differ noticeably in size from 
females. The largest specimen taken so far outside the Orient was taken in 
Aquatic Park, Berkeley, in March, 1970, measuring 185 mm. standard length 
and 234 mm. total length. 

The species is unusually tough and resilient. Five adult specimens were taken 
from brackish water (sodium chloride concentration unknown) off Napa Slough 
in San Pablo Bay in September, 1968, and unceremoniously dumped into fresh 
water. They survived, and 10 days later were transferred to pure sea water at 
24 hour intervals and in 20 percent increments. After 48 hours in salt water, they 
were transferred back to pure fresh water in the same manner. While not sub- 
ject to any systematic temperature manipulation, the specimens were kept in 
water of various salinities that ranged between 52° and 83° F. 

The dispersal of the mahaze up various freshwater river and canal systems 
appears less startling in view of the above facts. The goby obviously possesses 
the ability to penetrate up mud-bottomed lowland rivers; it commonly does this 



Vol. XXXVIII] BRITTAX ET AL.: ACANTHOGOBIUS FLAVIMANUS 213 

in the Orient and is doing it here. Acanthogobius jlavimanus has apparently 
been carried out of the Delta in the strong southward flowing current of the 
Delta-Mendota Canal, which has a distinct problem with silting and with the 
establishment of a Japanese freshwater mussel, which will favor the estabhsh- 
ment of such mud-associated bottom fish; its phenomenal increase in the San 
Luis Reservoir may be explained in terms of a habitat unoccupied by other 
bottom fishes (no catfishes were observed among the fish killed by the algal 
bloom and following die-off). The spreading of the species from San Francisco 
Bay to Bolinas Lagoon would be more difficult. While the mahaze can tolerate 
pure salt water, the shore between San Francisco Bay and Bolinas Lagoon is 
open rocky coast with a large assemblage of predaceous fishes. Either a migra- 
tion occurred in the face of severe ecological opposition, or transfer through 
human agency took place. Ocean-going ships do not enter Bolinas Lagoon, so 
direct transfer from the Orient is contra-indicated, though fishing boats and 
pleasure craft sometimes make the trip from San Francisco Bay. Further, .4. 
jlavimanus is being used as a bait fish to some extent in the Bay and Delta 
regions. Consequently, it is Hkely introduction into Bolinas Lagoon came about 
through discarding of bait fish or by tiny young fish carried out in discharged 
coolant water. It is not known whether the species is firmly established in Boli- 
nas Lagoon, as only 1 specimen (an adult) has been taken to date. 

Aquarium specimens remain on the bottom, burrowing in mud or sand but 
not in gravel. Swimming is accomplished by short jerks. Buccal respiration, or 
air gulping, was observed on 1 occasion, apparently initiated by an oxygen defi- 
ciency in the water. 

Virtually nothing is known of the ecology of the mahaze in California, al- 
though Okada (1960) gives some data for Japan. J. A. Aplin (personal commu- 
nication) observed that regurgitant intended to be fed by 3 great blue herons to 
their young in a rookery on Bair Island, one mile north of the port of Redwood 
City, in early August, 1969, consisted totally of several mahaze, the larger ones 
being about 8 inches total length. 

Besides Acanthogobius jlavimanus, 2 other exotic fish species have appeared 
in San Francisco Bay in the last 10 years or so through unknown means of intro- 
duction (Ruth, 1964, on information supplied by W. I. Follett). The rainwater 
fish, Lucania parva (Baird), a cyprinodont from brackish waters along the LT.S. 
Atlantic Coast, appeared first (Hubbs and Miller, 1965). Within a few years 
A. jlavimanus and another Oriental goby, Tridentiger trigonoccphalus (Gill), made 
their appearance. The latter, called "shimahaze" in Japan, is known from only 2 
localities around the Bay (1966), but is assumed to be firmly established; it is 
also recorded from Los Angeles Harbor (Hubbs and Miller, 1965). Brittan, Al- 
brecht, and Hopkirk (1963) give distinguishing characters for A. jlavimanus. 

In addition to the paper by Newman (1963) on the introduction into San 
Francisco Bay of an oriental commerical shrimp, Palacmon macrodactylus 



214 CALIFORNIA ACADEMY OF SCIENCES LProc. 4th Ser. 

(Rathbun), discussed by Brittan et al., (1963), a recent paper by Nijssen and 
Stock (1966) concerning the explosive spread of an eastern North American 
euryhaline amphipod, Gammarus tigrinus Sexton, in the Yssellake (or Ijsselmeer, 
the present name for the old, now smaller, Zuydersee), Netherlands should be of 
considerable value to those interested in such phenomena. Certainly more for- 
eign euryhaline fishes and invertebrates will show such sudden appearances, fol- 
lowed by rapid expansion of ranges, in estuarine systems the world over. 

The authors wish to thank the following persons for information and assist- 
ance: J. A. (Al) Aplin, Dr. Harold Chadwick, John M. Huddleson, and Robert 
McKechnie, California Department of Fish and Game; Dr. Earl Herald and 
Walter SchneebeH, Steinhart Aquarium, California Academy of Sciences; and 
Robert Hassur, Stanford University. 

LITERATURE CITED 

Brittan, M. R., A. B. Albrecht, and J. D. Hopkirk 

1963. An Oriental goby collected in the San Joaquin River Delta near Stockton, Califor- 
nia. California Fish and Game, vol. 49, no. 4, pp. 302-304, 1 fig. 
Dotu, Y., and S. Mito 

1955. On the breeding habits, larvae, and young of a goby, Acanthogohiiis jlavimanns 
(Temminck et Schlegel), Japan Journal of Ichthyology, vol. 4, nos. 4-6, pp. 
153-161, 5 text figs, (in Japanese, English summary). 
HuBBS, C. L., and R. R. Miller 

1965. Studies of cyprinodont fishes. XXII. Variation in Lucania parva, its establish- 

ment in western United States, and description of a new species from an inte- 
rior basin in Cohuila, Mexico. Miscellaneous Publications of the Museum of 
Zoology, University of Michigan, no. 127, 104 pp., 8 figs., 27 tables, 3 plates. 
Newman, W. A. 

1963. On the introduction of an edible Oriental shrimp (Caridea, Palaemonidae) to San 

Francisco Bay. Crustaceana, vol. 5, no. 2, pp. 119-132, 3 figs. 
Nijssen, H., and J. H. Stock 

1966. The amphipod Gammarus tigrinus Sexton, 1939, introduced in the Netherlands 

(Crustacea). Beaufortia, vol. 13, no. 160, pp. 197-206, 3 figs., 1 table. 
Okada, Y. 

1960. Studies on the freshwater fishes of Japan. Prefectural University of Mie, Tsu, 
Japan. 874 pp., 61 plates, 133 figs., 135 tables ("1959-1960" on title page). 
Ruth, F. S. 

1964. Habitat check list of the vertebrates of the San Francisco Bay Region of Califor- 

nia. 16 pp. Mimeotype Corporation, Walnut Creek, CaHfornia. 



PROCEEDINGS 

OF THE 

CALIFORNIA ACADEMY OF SCIENCES 

FOURTH SERIES 



Festschrift for George Sprague Myers 



Vol. XXXVIII, No. 12, pp. 215-264; 20 figs. December 31, 1970 



SOME NERVE PATTERNS AND THEIR 

SYSTEMATIC SIGNIFICANCE IN PARACAN- 

THOPTERYGIAN, SALMONIFORM, GOBIOID, 

AND APOGONID FISHES -^ 

By 

Warren C. Freihofer 
Division of Systematic Biology, Stanford University 

Some time ago while examining the salmopercoid genera Percopsis and Aphre- 
doderus for the ramus laterahs accessorius, I noticed from superficial dissections 
on the head that these fishes had some strikingly interesting nerves emanating 
from one main point in the upper cheek region next to the preopercle (Freihofer, 
1960). The nerves came up to the skin from their source below on the truncus 
hyomandibularis. There were four main nerves, one to each of the roofing mem- 
branes of the supraorbital, infraorbital, preopercular, and mandibular canals. In 
1950 Ray had reported similar nerves for the lantern fish Lampanyctus leu- 
copsarus and recognized them to constitute a group of nerves which she called 
the ramus canalis lateralis facialis system. The statement by Frost (1926) that 
the otoliths of Apogon and of the salmopercoid fishes strongly resembled each 
other led me to examine apogonids for these ramus canalis nerves. They were 
found to have these nerves in similar pattern. These facts suggested that there 
might be a relationship between lantern fishes, salmopercoid (or percopsiform 
fishes, as they are now called), and the supposedly percoid apogonids. 



1 Research for this paper was supported by National Science Foundation Grant GB-198. 

2 A summary of results was kindly read for me by Dr. G. S. Myers at the New York City meetings of the 
American Society of Ichthyologists and Herpetologists in June, 1969. 

3 Submitted for publication February 2, 1969. 

[215] 



216 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

The salmopercoid fishes are especially intriguing since they obviously appear 
to combine features of both salmonids and percoids. Detailed studies of the ner- 
vous and skeletal systems of salmopercoids were begun as well as a survey of 
many fish groups for the ramus canalis lateralis system and the ramus lateralis 
accessorius. Two events influenced the survey. One was the publication in 1966 
by Greenwood, Rosen, Weitzman, and Myers of a classification in which a new 
superorder of fishes, the Paracanthopterygii, was proposed which brought to- 
gether six orders or parts of orders, some for the first time. The survey was 
directed to include all these groups. Its success was furthered by a second event, 
my fortuitously being present on cruise 16 of the R/V Anton Bruun. Bottom 
hauls frequently brought up large numbers of most major groups of paracanthop- 
terygian fishes, thus affording material that could be processed in the Sihler 
technique. Preliminary observations of the nerves of all these paracanthop- 
terygian fishes might, I thought, reveal nerve features that would test the va- 
lidity of this new superorder as well as help in the question of codfish and bro- 
tulid relationships, fishes which looked much alike as they lay together on the 
ship's deck. A simple soul, but not perhaps an "educated" one, would think 
these latter two groups must be related. To suppose that the percopsids, batra- 
choidids, ogcocephalids, gobiosocids, lophiids, and their related families, had a 
relationship to each other let alone to the codfishes, hakes, ophidioids, and 
zoarcids would have and apparently still does strain beyond the bounds of belief 
the minds of most ichthyologists. The results of the nerve survey should lessen 
doubts now generally held about the Paracanthopterygii. 

There is enough detail in the descriptive section of patterns of the ramus 
lateralis accessorius (RLA) to show how similar these are in brotulid, ophidiid, 
and gadiform fishes. Preliminary information is given on a special enlarged lat- 
eral-line branch (or several branches) which supply the pectoral-pelvic area in 
percopsiform, brotulid, batrachoid, and gobioid fishes and which is part of a seg- 
mental series of lateral-line nerves. This nerve and the segmental series of which 
it is a part may be a primitive paracanthopterygian feature inherited from lower 
fishes. A preliminary account is given of the manner in which the fin-ray nerves 
course in paracanthopterygian fishes in contrast to the way they do in acan- 
thopterygian and numerous other fishes. It is a feature that appears character- 
istic but not unique to the Paracanthopterygii. 

Brief, preliminary, comparative studies of the ramus canalis lateralis system 
of nerves are given for a number of families. A much fuller treatment is planned 
for a future paper. 

One of the purposes of the present paper is to report some features of nerves 
which bear on the systematic validity of the Paracanthopterygii. Other main sys- 
tematic questions to which the results presented are relevant are: (1) relation- 
ships of the salmopercoid fishes to other paracanthopterygian fishes ; ( 2 ) the re- 



Vol. XXXVIII] FREIHOFER: XERVE PATTERNS 217 

lationships of salmopercoid fishes to acanthopterygian fishes; (3) the origin of 
the salmopercoid fishes and of gadoid, ophidioid and batrachoid fishes; (4) the 
interrelationships of the gadoid and ophidioid fishes; (5) the relationships and 
reclassification of the gobioid fishes; (6) the relationships of the Apogonidae. 

A note of explanation on the ramus lateralis accessorius. It supplies taste buds 
on the body and or fins. It is not a lateral-line nerve. The name ramus recurrens 
facialis is more descriptive and can be shortened to "recurrent facial," but the ab- 
breviation "RLA" for the former term is used here. 

MATERIALS AND METHODS 

Specimens of Merluccius gayi, Brotula clarkae, Physkulus talarae, Lcpophi- 
dium prorates, Porichthys margaritatus, Bathygobius lineatus, Hoplostethus paci- 
jicus, Melamphaes species, Apogon astradorsatus, Scopelengys tristis, and Zali- 
cutes elater were collected on cruise 16 of the R/V Anton Bruun and specimens of 
Brotuloides enimalas by Margaret Bradbury on cruise 19 of the R/V Te Vega. 
All were kept in formalin until processed for the nerves by the Sihler technique 
(Freihofer, 1966), in which the stained nerves stand out in transparent whole 
specimens. Alizarin specimens were also prepared for examination of the skeleton. 
Both Sihler and alizarin preparations were also made of Percopsis omiscomaycus, 
Percopsis transmontana, Aphredoderus say anus, Hypomesus pretiosus, H. olidus, 
Dkrolene intronigra. The following specimens were also examined by dissection 
under the microscope: Chologaster papillijerus and Dkrolene kanazawi, uncata- 
logued; Lamprogrammis niger, LACM (Los Angeles County Museum) 9708-5, 
Watasea sivicola, 26797; Monomerepus species, 57024; Dinematkhthys iluo- 
coeteoides, uncatalogued; Eutyx diagrammus uncatalogued; Merluccius produc- 
tus, LACM 9815-8; Eleginus gracilus, 49233; Mkrogadus proximus, 49237; 
Boreogadiis saida, 4S810; Urophycis jloridanus, 50878; Laemonema barbatulum, 
63261; Coelorhynchus scaphopsis, 179; Eleotris juscus,\\\\cdiia\ogwed. Numerous 
other species of various families and orders were examined but are not listed. All 
catalogue numbers are from the Stanford University fish collections unless other- 
wise noted. 

Many of the families and genera mentioned in the text were also examined 
from specimens in the Starks skeletal collection at Stanford University. 

ACKNOWLEDGMENTS 

I would like to express my appreciation for the generous and essential help 
the following people gave in sending specimens kept specially in formalin or who 
helped in the field: Carl Bond, Oregon State LTniversity; Margaret Bradbury, 
San Francisco State College; Daniel M. Cohen, Systematics Laboratory, U. S. 
Fish and Wildlife Service; E. J. Crossman, Royal Ontario Museum; William 
Eschmeyer, California Academy of Sciences; Carter Gilbert, Florida State Mu- 



218 CALIFORNIA ACADEMY OF SCIENCES LProc. 4th Ser. 

seum; Shelly Johnson, University of Southern California; Robert Lea, California 
Division of Fish and Game; the late Fr. Romeo O. Legault, University of Ottawa; 
Donald McPhail, University of British Columbia, John Massie, California Aque- 
duct Facility; William Weaver, Florida State Museum; James Davis, North 
Carolina, State Fish and Game. 

Observations on osmerids are from Sihler preparations and serial sections 
made by Craig Findly as part of a special problems study he was doing at Stan- 
ford University. 

Dr. Daniel Cohen identified the gadoid and ophidioid fishes collected on the 
R/V Anton Bruun except for Lepophidiuvi prorates identified by Dr. C. Richard 
Robins; Dr. Ernest Lachner identified the apogonid and Doug Hoese the goby. 

My special thanks are due Adair Fehlmann, Smithsonian Oceanographic 
Sorting Center, for arranging my participation on cruise 16 of the R/V Anton 
Bruun and for much additional help in procuring specimens. Leonard Compagno 
was a patient listener and excellent discussant while the work was in progress. 
Max Millsap, Stanford Anatomy Department, took all photographs except that 
in figure 17 which was taken by the author. 

RAMUS LATERALIS ACCESSORIUS 

Description of Ramus Lateralis Accessorius in the Ophidiidae 

In Lepophidium prorates an enormous ramus lateralis accessorius (RLA) 
arises from the geniculate ganglion (fig. 1), passes dorsolaterally and posteriorly 
up to the cranial roof where it bifurcates into a large branch (RLA-PP) going to 
the pectoral and pelvic fins and a small branch (RLA-PDA), about % the size of 
the other, which goes to the dorsal and anal fins. The pectoral-pelvic branch turns 
posterolaterally beneath the cranial roof, enters an intraosseous passageway in the 
parietal, leaving it at its posterolateral corner by a large foramen. 

The dorsal-anal branch, RLA-PDA, after leaving the parietal, passes poste- 
riorly near the middorsal line, beneath skin back to the dorsal fin where it dips 
ventrally and passes posteriorly alongside the pterygiophores about % of their 
length below their outer distal ends. It forms a longitudinal plexus with crossing 
branches of the dorsal rami of the spinal nerves supplying the fin rays and mem- 
brane. At the seventh and eighth dorsal crossing segmental rami, 2 large branches 
are given off, one at each of these segments, which pass beneath the skin postero- 
ventrally towards the origin of the anal fin. These 2 branches of RLA-PDA join 
at the second segment from the anal origin and pass inwardly and run posteriorly 
alongside the pterygiophores forming a longitudinal plexus with branches of the 
crossing ventral spinal rami supplying the fin rays and membrane. On their course 
from the dorsal to the anal fins the 2 branches of RLA-A cross numerous 
branches of the lateral line and segmental rami going to the skin and also exchange 
a few branches between each other. The nerves from the longitudinal plexi of 



Vol. XXXVIII] 



FREIHOFER: NERVE PATTERNS 



219 




220 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

the dorsal and anal fins enter the fin rays and membrane in a characteristic way. 
For each segment, a branch of RLA serves the half fin ray of its side and its half 
membrane lying posterior to the half ray. In doing so, the fin-ray nerve passes 
inward towards the opening in the base of the ray and gives off a branch which 
runs distally in the fin membrane of its side of the body. The rest of the fin-ray 
nerve passes forward into and through the split base of the fin ray and then 
onto the outer, external surface of the half of the ray of its side and along the 
half fin ray to its distal end. The longitudinal plexus of each side of the dorsal 
fin continues posteriorly and meets its counterpart from the anal fin at the mid 
point of the tip of the hypural fan. 

Emerging from the posterolateral corner of the parietal, the pectoral-pelvic 
branch, RLA-PP, passes across the medial surfaces of the supratemporal canal 
bone and the epiotic arm of the posttemporal and ventrally beneath skin along 
the posterior edge of the supracleithrum and cleithrum en route to the pectoral 
and pelvic fins. A short distance above the pectoral fin a branch is detached 
which passes along the dorsal edge of the muscular base of the pectoral fin. A 
branch from the branchial plexus joins the pectoral branch of RLA and the com- 
pound nerve enters the bases of the fin rays ventrally, giving off branches to the 
half rays and membrane. As for the fin rays of the other fins, the nerves to each 
half ray course on the outside surface of the rays, not internally between each of 
the halves of each fin ray as is characteristic of percoid and most other fishes. 
As RLA-PP passes the ventral edge of the pectoral fin a branch is detached 
which joins a nerve from the brachial plexus which enters the ventral base of the 
pectoral fin and passes dorsally up through the pectoral fin giving off branches 
to each fin ray as described for the dorsal base of the pectoral fin. 

The enormous remainder of RLA-PP passes anteroventrally beyond the 
pectoral fin to enter the pelvic fin. At the posterior end of the fleshy isthmus 
between the two gill openings, the large trunk of RLA-PP of each side of the 
body join in the midventral line and continue anteriorly as one trunk. Next, the 
common trunk is joined by a large spinal ramus of each side that comes to the 
surface at the midventral line after having passed down the medial side of the 
body wall. The resulting huge common spinal and recurrent facial trunk divides 
at the base of the pelvic fin. A branch is given to each of the two pelvic fin-ray 
bases. Each pelvic fin ray has the fin-ray nerve coursing on its external surface 
as in the other fins, a pattern which, as has been mentioned, is significantly dif- 
ferent from that for percoid and many other fishes. 

Description of the Ramus Lateralis 
AccESSORius IN the Brotulidae 

In Ogilbia ventralis, studied from a dissected alcoholic specimen only, a very 
large RLA arises from the geniculate ganglion (fig. 2) and passes a rather long 



Vol. XXXVIII] 



FREIHOFER: NERVE PATTERNS 



221 




(IH 



222 CALIFORNIA ACADEMY OF SCIENCES LProc. 4th Ser. 

distance posterodorsally up to the cranial roof where it is met by an extremely 
thin vagal ramus just before cranial exit through the parietal near its postero- 
medial corner. Immediately outside the cranium on the nape it divides into al- 
most equal sized branches, RLA-PDA to the dorsal and anal fins, and RLA-PP 
to the pectoral and pelvic fins. Branch RLA-PP to the pectoral and pelvic fins 
passes posteroventrally following the posterior edges of the posttemporal, supra- 
cleithrum and cleithrum directly beneath skin. It passes medially behind the 
dorsoposterior end of the cleithrum and, upon emerging from its posterior edge, 
it detaches a large branch along the dorsal edge of the muscular base of the pec- 
toral fin. This branch is joined by a large spinal nerve branch of the brachial 
plexus just before entering the upper end of the base of the pectoral fin. The 
remainder continues ventrally posterior to the cleithrum and beneath skin on 
its way to the pelvic fin where it is met at the midventral line by a large branch 
of a spinal nerve. In passing the ventral end of the pectoral fin, a small branch 
of RLA passes dorsoposteriorly to the ventral end of the pectoral fin base where 
it joins a branch of the brachial plexus that enters the ventral end of pectoral 
fin base. 

Branch RLA-PDA runs along the base of the dorsal fin as in other brotulids. 
A large branch detaches from the dorsal fin branch at about the eleventh seg- 
ment back from the anterior end of the dorsal fin. A second fair-sized branch 
arises at about segment 22. 

With its 2 main branches arising from the main trunk of RLA outside of the 
parietal, the pattern in Ogilbia is basically like that in the gadiform fishes. 

The same pattern was found in Dinematkhthys iluocoeteoides. 

In Brotula clarkae (fig. 3), the pattern of RLA is the same as that described 
for Brotula multibarbata from a dissection of an alcoholic specimen (Freihofer, 
1963). An anal branch was not noted at that time. A Sihler preparation of 
Brotula clarkae shows that there is an anal branch similar to that in the ophidiid 
Lepophidium and the morid, Physiculus. 

The branch to the dorsal and anal fins, RLA-PDA, passes into the trigeminal 
foramen with the supra- and infraorbital trunks, but inside the foraminal pas- 
sageway of the prootic bone RLA-PDA departs and passes dorsoposteriorly 
through the body of the sphenotic and re-enters the cranium where it continues 
dorsoposteriorly over the cranial ceiling up to its exit via an osseous passageway 
in the parietal at its posteromedial corner next to the base of the supraoccipital 
spine. On one Sihler preparation 5 anal branches are detached from RLA-PDA. 
The first two, which join and separate again on their way to the anal fin, come off 
at segments 9 and 10 counting from the origin of the dorsal fin. The next three 
come off at segments 13, 14, and 19. These branches enter the anal fin respec- 
tively at 2, 7, 9, 10, and 19 segments from its origin. Four similar branches occur 
on the other side. Two other specimens had 4 anal branches similarly spaced. Fi- 



Vol. XXXVIII] 



FREIHOFER: XERVE PATTERNS 



223 




m 



224 CALIFORNIA ACADEMY OF SCIENCES IProc. 4th Ser. 

nail)', a fourth specimen had 1 large anterior anal branch entering the anal fin 
at the twentieth ray with no RLA innervation going to these first 20 rays, ap- 
parently, and with a second anal branch in the position of the last branch on the 
other specimens. 

The pectoral-pelvic branch, RLA-THYO, leaves the cranium on the postero- 
medial surface of the truncus hyomandibularis, turns abruptly posteriorly, and 
runs along the medial side of the hyomandibular bone (fig. 3). Emerging from 
behind the posterior end of the hyomandibular, it passes dorsal to a large oper- 
cular muscle, then under the long pterotic spine, and onto the medial surface 
of the posttemporal. It runs for a short distance beneath skin between the 
posttemporal-supracleithral articulation then medially again behind the cleith- 
rum, coming from behind this bone above the dorsal end of the muscular base 
of the pectoral fin where a fair-sized branch is detached which runs out the 
dorsal edge of the muscular pectoral base, being joined en route by a larger 
branch from the brachial plexus. The dorsal pectoral branch of RLA-THYO 
detaches 2 small nerves before joining the brachial branch. One of these goes 
to skin near the base of the fin rays on the medial side of the fin. The other goes 
to skin on the lateral side. Another branch to the lateral surface of the fin base 
is detached halfway down the fin. As RLA-THYO passes beneath skin near the 
ventral end, a large branch is detached which, together with a large branch from 
the brachial plexus, enters the ventral base of the fin rays. The dorsal and ven- 
tral fin-ray nerves become smaller as they approach each other as they pass 
through the bases of the fin rays giving off a branch between each 2 succeeding 
rays, the fin-ray branch of each side of the fin coursing distally in the membrane 
between each succeeding whole ray. The main trunk of RLA-THYO passes an- 
teriorly beneath the skin paralleling the cleithrum. Near the base of the pelvic 
fin it joins with a large branch of a ventral spinal nerve, which, before joining 
RLA-THYO, gives off a branch to the pelvic muscles. Before joining with the 
branch from the brachial plexus, RLA-THYO detaches a branch which goes to 
the anterior surface of the first ray. It is joined by a branch from the ventral 
spinal ramus. The posterior external surface of the first ray plus the anterior 
and posterior surfaces of the second pelvic ray are innervated by 2 compound 
branches from RLA-THYO and spinal nerve trunk. In reaching the pelvic fin 
the ventral spinal nerve has passed down the wall of the body cavity medial to 
the pectoral girdle and comes to the surface at the midventral line near the base 
of the pelvic fins. 

Description of RLA in Gadiform Fishes 

In the morid Physiculus talarae a large RLA leaves the geniculate ganglion, 
courses dorsally in an open groove on the inner surface of the parietal, exits 
through a large foramen in this bone near its center and passes posterolaterally 
beneath the large muscle mass on top of the head and then dorsally between the 



Vol. XXXVIII] 



FREIHOFER: NERVE PATTERNS 



225 




226 CALIFORNIA ACADEMY OF SCIENCES LProc. 4th Ser. 

side of the muscle mass and the medial surface of the supratemporal canal bone. 
A thin vagal ramus joins RLA at its exit through the parietal. Medial to the 
supratemporal canal RLA bifurcates into RLA-PDA, the dorsal-anal branch, 
and RLA-PP, the pectoral-pelvic branch. RLA-PDA curves medially beneath 
the skin on its course to the dorsal fin at the anterior end of which it detaches 
an anal fin branch, RLA-A, which on one specimen parallels the course of 
RLA-D for 3 segments and then curves ventroposteriorly under the skin. On 
another specimen the dorsal branch did not detach the first anal branch, a small 
one to the anterior part of the anal fin, until the ninth segment from the origin 
of the dorsal fin. The main large anal branch detaches on this specimen at the 
twelfth segment. On the specimen illustrated, the anal branch passes beneath 
the skin and divides en route into 3 parts (fig. 4), a smaller anterior branch 
which reaches the anal fin at the sixth segment from its origin and forms an 
anteriorly coursing longitudinal plexus with crossing ramuli of the segmental 
ventral spinal nerves. The second or main part of the three reaches the base of 
the anal fin at its fifteenth ray and forms a posteriorly coursing longitudinal 
plexus. The third branch joins the longitudinal plexus at the twenty-first ray. 
The plexus continues all the way to the caudal fin. On the other side of the 
specimen 2 anal branches course beneath the skin paralleling each other and 
joining at the base of the anal at about its fifteenth ray. 

The branches of the longitudinal plexus that enter the fin rays do so in nearly 
the same way that they do in the ophidiid, Lepophidium. In Physiculus a fin- 
ray nerve passes to the posterior side of the base of each half fin ray of its side 
of the body. Before passing anteriorly through the opening between the two di- 
verging bases of the half fin rays each fin-ray nerve of each side detaches a 
branch that runs distally out the posterior external surface of the fin ray. The 
remainder of the fin-ray nerve of each side passes through the opening between 
the base of the half fin rays and passes distally on the external anterior surface 
of the half fin ray. As was mentioned for the ophidiid Lepophidium, this is dras- 
tically different from the pattern shown in most other fishes, for which Aphre- 
doderus (fig. 13) is an example, where the fin-ray nerves course in the hollow 
internal tube formed by the two concave half rays of each side meeting. 

From its origin from the main trunk of RLA, the pectoral-pelvic branch, 
RLA-PP, passes posterolaterally on the medial surface of the supratemporal 
canal, crosses laterally over the epiotic arm of the posttemporal and ventrally 
beneath the skin slightly posterior to the supracleithrum and cleithrum but an- 
terior to the pectoral fin and lateral to the base of the fin. Dorsal to the pectoral 
fin base a branch is detached from RLA-PP which joins with a large branch 
from the brachial plexus which together enter as one nerve the dorsal base of the 
pectoral fin and pass down through the basal opening between the fin-ray halves 
giving off a branch to each half fin ray as already described for the dorsal and 
anal fins. Further on its course ventrally over the lateral surface of the muscular 



Vol. XXXMII] FREIHOFER: XERVE PATTERXS 227 

base of the pectoral fin, a smaller branch detaches from RLA-PP and passes 
over towards the ventral end of the pectoral fin base where it is met by a large 
nerve from the brachial plexus. The compound nerve enters the ventral end of 
the base of the pectoral fin and passes dorsally through successive rays until it 
meets with the branch coming from the dorsal end of the pectoral base. 

The very large remainder of RLA-PP passes ventrally beneath the skin to the 
slightly jugular pelvic fin where it is met by a large branch from the first ventral 
spinal nerve posterior to the occipito-spinal complex. The fin-ray nerves run 
distally on the external surface of the fin rays as already described for the other 
fins. 

In the Merlucciidae, no RLA was found in a Sihler nerve preparation of 
Merluccius productus. 

The pattern in the Gadidae as represented by Microgadus (fig. 5) differs 
significantly from that in the Moridae in that the anal-fin branch in the Gadidae 
detaches from the pectoral-pelvic branch not far above the base of the pectoral 
fin, whereas in the INIoridae the anal branch detaches from the dorsal fin branch 
about a dozen segments back from the origin of the dorsal fin or by one or two ad- 
ditional branches more posteriorly. 

Other gadid genera examined showing the same pattern as Microgadus are 
Gadus, Boreogadus, and Eleginus. 

In the Macruridae the dorsal-anal branch detaches after the main trunk of 
RLA passes the epiotic arm of the posttemporal. This pattern bears a small 
resemblance to the gadid pattern but a larger one to the morid pattern. In 
Macrurus RLA crosses laterally the posttemporal bone and divides, one branch 
passing down the posterior edge of the supracleithrum and cleithrum to supply 
the pectoral and pelvic fins, and the other arching dorsally then straightening 
out, crossing several septa and then sending one branch up to the overlying 
dorsal fin and one ventroposteriorly to the anal fin. 

A branch of a dorsal ramus to the first elongated spine of the dorsal fin was 
found which is of unusual interest for the speculation that it arouses as to its use 
in the life of macrurids. All the numerous species examined had it well developed 
but not as conspicuously developed as in Lionurus gibber. In this species a rela- 
tively enormous nerve extends out in the posterior groove of the elongated an- 
terior dorsal spine and out beyond the spine in the protective sheath that is more 
than once again as long as the spine. This projection appears to be largely 
nerve. No enlarged nerve was observed in any of the other fins. For a sup- 
posedly deep-sea bottom swimming fish, this great dorsally directed tactile and 
taste feeler is surprising. As was mentioned, this development is much greater 
than anything seen on the ventral or lateral fins. Some important stimuli from 
above the fish must be perceived. It suggests that the elongated nerve-filled 
spine is a contact organ for use in touching other individuals swimming directly 
above in schooling. 



228 



CALIFORNIA ACADEMY OF SCIENCES 



IProc. 4th Ser. 




!^ 



Vol. XXXVIII] FREIHOFER: SERVE PATTERNS 229 

Pattern of RLA in Percopsiformes 

In Pcrcopsis (fig. 6) the sympathetic chain and RLA course bound together, 
the two forming a common trunk. All the branches that are given off from this 
common trunk may contain fibers from both these nerve trunks. The RLA- 
sympathetic common trunk is very large, 4 times the size of the ninth cranial 
nerve. The sympathetic chain alone in most fishes is usually less than ^/n the 
size of the ninth cranial nerve. 

The RLA part of the common trunk arises from the geniculate ganglion. 
What evidently is a sympathetic trunk, and which is about %o the size of the 
RLA-sympathetic common trunk, curves anteromedially around the root of the 
truncus hyomandibularis. It was not followed further on the surface of the 
truncus hyomandibularis as this truncus and the truncus infraorbitalis come to- 
gether as they pass medially towards the fifth-seventh complex. The common 
trunk of RLA + sympathetic chain leaves the truncus hyomandibularis shortly 
outside the cranium and passes posteromedially across the wall of the otic bulla, 
crossing the ninth cranial nerve en route to which it is connected by a thin 
branch which evidently is sympathetic. The common trunk passes onto the 
ventral side of the vertebral column and then along it to the caudal fin. As the 
common trunk passes the pharyngo-branchial roots of the vagus, 2 sympathetic 
branches leave from 2 ganglia located on the surface of the common trunk. One 
of the branches is about Mo the size of the common trunk; the other is about ¥20 
its size. At the same point a large branch, about Vi the size of the common trunk 
leaves from a bundle of fibers already formed anterior to the 2 sympathetic 
ganglia. This large branch must be mostly RLA fibers. It does not come from 
these 2 ganglia. This large branch passes to the base of the first and second ven- 
tral spinal rami of the occipito-spinal nerves. Opposite Baudelot's ligament a 
small branch detaches from the common trunk, runs parallel to the ligament, 
and then joins the third ventral occipito-spinal ramus. The next branch is a large 
one, about V-i the size of the first large trunk branch. It goes to the fourth ven- 
tral spinal ramus. The next branch is about '''^ as large as this one and each 
succeeding branch going to each ventral spinal ramus is small. The specimen 
illustrated in figure 6 is different from the specimen described in the text. 

The Amblyopsidae have the same pattern as Percopsis, not as in Aphrcdo- 
dcrus. In Chologastcr papilliferus RLA appears even larger. 

In Aphrcdoderiis sayanus (fig. 7) the pattern is quite different from that 
in Percopsis, but one basic similarity remains, that of having the fibers of RLx^ 
destined for the pectoral fins, and perhaps for the pelvics also, distributed via 
the occipito-spinal complex. 

In Aphredodcrus a sizable RLA leaves the cranium through the parietal and 
courses beneath the skin towards the dorsal fin. Directly outside the cranium 
RI.A detaches a branch down the posterior surface of the cranium that joins the 



230 



CALIFORNIA ACADEMY OF SCIENCES 



[Proc. 4th Ser. 



o 


(■5 




z 


o 


00 




CO 


a. 


Z 


oc 


> 


CO 


a 


CO 




o 
o 


6 


> 




O 


o 






oc 


O^ 






Q 




~\\ 






« 2 



Vol. XXXVIII] 



FREIHOFER: SERVE PATTERNS 



231 






o 


Oh 




o 






CL 


t^ 


o 


CO 


u 


o 


6 


« 


a. 


o 


o 


(/) 


o 


b 


6 


cc 




o 


i 




o 






cc 






a 







232 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

occipito-spinal complex where the first dorsal ramus leaves its vertebra. Varying 
with the specimen, this first branch of RLA either courses to the first occipito- 
spinal nerve independently of the first dorsal ramus which passes up the pos- 
terior surface of the cranium and innervates skin overlying the posterolateral 
area of the cranium; or the two run in common, or part of the way in common, 
but in opposite directions. As RLA continues towards the dorsal fin other dorsal 
spinal rami cross it. It appears that a second branch from RLA courses down the 
posterior edge of the first neural spine to join another part of the occipito-spinal 
complex. The crossing dorsal rami and RLA form a longitudinal plexus alongside 
the dorsal pterygiophores. The plexus continues beyond the dorsal fin but in 
diminished size until it reaches the caudal fin. It could not be determined if any 
RLA fibers remained in it that far. No branch of RLA to the anal fin was dis- 
cernible nor was there a pronounced longitudinal plexus along the anal as there 
is along the dorsal fin. 

The common trunk of RLA + sympathetic chain in Pcrcopsis is about 20 
times the size of the sympathetic trunk in Aphredoderus where this trunk crosses 
the wall of the otic bulla. 

Pattern of RLA in Batrachoididae 

In Porichthys niargaritatus (fig. 8) RLA must be looked for in the same 
place as in Percopsis, that is, issuing from the facial foramen together with the 
tr uncus hyomandibularis combined with the sympathetic chain. Shortly outside 
of the facial foramen the common trunk of RLA + sympathetic chain leaves 
the truncus hyomandibularis, turns posteriorly and separates, the sympathetic 
trunk passing posteromedially across the wall of the otic bulla and onto the 
ventral side of the vertebral column and along it to the caudal fin. The trunk 
of RLA passes posterolaterally across the wall of the otic bulla, gradually di- 
verging from the sympathetic trunk. RLA crosses the ninth cranial nerve, drops 
ventrally, is pierced by a branch of the vagal trunk with no exchange of fibers. 
This vagal branch passes directly laterally, then ventrally across the base of the 
opercular spine and innervates skin down as far as the branchiostegal rays. As 
the main trunk of RLA passes posteriorly medial to the cleithrum, the pectoral- 
dorsal and pelvic-anal branches arise (fig. 8). The pectoral-dorsal branch courses 
laterally, reaching the skin directly posterior to the cleithrum from which point 
it arches dorsoposteriorly towards the dorsal fin. A branch is given off which 
passes to the dorsal end of the pectoral fin base. The pelvic-anal branch crosses 
the medial side of the cleithrum and passes down the body wall medial to the 
base of the pectoral fin, dividing shortly beyond the cleithrum into the pelvic 
and anal branches. The anal branch slants towards the anal fin. The pelvic 
branch turns anteroventrally at the tip of the postcleithrum and courses beneath 
skin towards the base of the pelvic fin which it enters. A more complete de- 
scription will be published elsewhere. 



Vol. XXXVIII] 



FREIHOFER: yERVE PATTERNS 



233 




234 



CALIFORNIA ACADEMY OF SCIENCES IProc. 4th Ser. 




Figure 9. Dicrolene intronigra (Brotulidae) showing pectoral accessory lateral-line nerve 
(PP-ACC-LAT) and several succeeding ventral segmental accessory lateral-line branches. 
Nerve endings for scattered free lateralis organs visible as relatively short horizontal lines 
extending towards left (posteriorly) on six secondary branches of the ventral lateral-line 
nerves. Sihler preparation. 



Pectoral-Pelvic Accessory Lateral-Line Nerves in 
Percopsiform, Gadoid, Ophidioid, Batrachoidid and Gobioid Fishes 

Accessory, ventrally directed, segmental lateral-line nerves are poorly known 
in other fishes. They were known previously apparently only for the hatchet- 
fish, Argyropelecus (Handrick, 1901). The account given here is brief; an ex- 
tended treatment is to be published later. In Percopsis as in Aphredoderus (fig. 
10, Pec-Pel ACC) a large branch is detached from the base of the ramus dorsahs 
of the main lateral-line nerve directly past the supracleithrum, the branch pas- 
sing in the skin paralleling the cleithrum, curving anteriorly around the ventral 
end of the base of the pectoral fin, and sending out branches from where it be- 
gins to curve and on all the way around below the base of the fin onto the skin 
of the anterior surface of the muscular base of the pectoral fin. The branches 
radiate out towards the midventral line and the base of the pelvic fin. Two seg- 
ments further posteriorly, there is another sizable but much smaller branch from 



Vol. XXXVIII] 



FREIHOFER: NERVE PATTERNS 



235 




PELVIC FIN 



Figure 10. Pectoral-pelvic area of Aphredoderus sayanus showing pectoral-pelvic ac- 
cessory ventral lateral-line nerve (PEC-PEL. ACC). Another nerve appears connected to the 
main lateral-line nerve close to PEC-PEL. ACC. but it is a cut spinal nerve. Sihler preparation. 



the ramus dorsalis of the lateral-line nerve. It passes ventrally, parallel to the 
large pectoral-pelvic accessory lateralis nerve but this smaller one serves mostly 
only skin at the ventral end of its segment. In each succeeding segment there is 
a similar but much smaller branch running ventrally. These lateralis branches 
innervate free lateralis organs scattered in the skin which are especially abundant 
in the area below and in front of the pectoral base and back to the pelvic base 
(fig. 11). The organs are borne on the scales. 

The same large accessory lateralis nerve is present in Aphredoderus. 

In the gadoid Merluccius gayi the segmental lateral-line nerves are present 
and in the same pattern as for Percopsis except that the anterior nerves are not 
enlarged. 

In the brotulid Dicrolene intronigra (fig. 9, PP-ACC-LAT), in addition to 
the large first pectoral-pelvic accessory lateralis nerve, there is a second large 
one in the following segment and then 3 more smaller ventral lateralis nerves 
at about trunk segments 5, 8, and 14 and at least 4 more on the remainder of 
the trunk. The first and largest has a verv similar course and distribution to 



236 



CALIFORNIA ACADEMY OF SCIENCES [Proc. 4Tn Ser. 




PEC -PEL 



Figure 11. Detail of pectoral-pelvic accessory lateralis nerve breaking up into smaller 
branches below base of pectoral fin in Aphredoderus sayanus. Sihler preparation. 



the first one in Percopsis except that it arises from the main trunk of the lateral- 
line nerve, not from the ramus dorsalis. It supplies large lateralis organs scat- 
tered in the skin ventral and anterior to the pectoral fin and up to the jugular 
pelvics. The second large branch passes ventroposteriorly and divides about 
halfway to the midventral line. The branches diverge supplying large lateralis 
organs in skin on the ventral body surface (fig. 9). The other branches start 
towards the anal fin and run horizontally at the ventral longitudinal septum 
supplying lateralis organs near this line. 

Other brotulid genera examined showing similar pectoral-pelvic and ventral 
accessory lateralis nerves are Bassozetus, Monomitopus, Monornerepus, and 
Porogadus. 

In the batracoidid Porkhthys margaritatus a pectoral-pelvic accessory lateral- 
line nerve arises from the deep main trunk of the lateral-line nerve as this nerve 
approaches medial to the supracleithrum. Further ventrally the pectoral-pelvic 
accessory lateral-Hne nerve divides into a large branch which drops ventrally 
close to the posterior edge of the postcleithrum, curves anteriorly at the ventral 
end of the pectoral base, and supplies the ventral lateral line from a little pos- 



Vol. XXXVIII] FREIHOFER: .\ERVE PATTERNS 237 

terior of the pectoral fin all the way to the most anterior extent of the ventral 
lateral line. The other smaller branch from the main pectoral-pelvic accessory 
lateral-line nerve passes ventroposteriorly and shortly divides into a branch that 
drops down to supply a segment of the ventral lateral line posterior to the pec- 
toral fin and a branch which continues posteroventrally, joining consecutively 
with 2 small and 1 large branch all separated from each other by some distance 
where they come off the main deep lateral-line nerve. These fused longitudinal 
branches supply the ventral lateral hne to the end of the abdominal area where 
additional branches from the main lateral-line nerve continue supplying the post- 
abdominal part of the ventral lateral line, these branches passing laterally out 
the horizontal septum independently of spinal rami. They leave from the deep 
lateral-line nerve located at the side of the vertebral column. 

The ventral lateralis organs and nerves of Dicrolene and Porichthys are a 
very noteworthy similarity between these fishes. 

In the gobies Bathygoblus lineatus and Acanthogobius jlavimanus, there is 
an accessory pectoral lateral-line nerve. In Bathygobius it detaches from the 
main lateral line midway between the location of the second and third ribs and 
courses ventrally close to the anterior edge of the third rib. The accessory pec- 
toral branch innervates a row of about a dozen free neuromasts which ends at 
the ventral edge of the pectoral fin but 2 segments posterior to it. No other 
ventral segmental accessory lateral line branches were observed. 

For the Gonostomatidae {Gonostoma elongatwn) and Chauliodontidae {Chau- 
liodus macouni) segmental ventral accessory lateral-line nerves extend to near 
the midventral line. The second such nerve is the largest and passes around 
the base of the pectoral fin. A single large accessory pectoral lateral-line nerve 
was reported for Argyropelecus (Sternoptychidae) by Handrick (1901). In the 
Osmeridae {Hypomesus pretiosus) a ventral segmental lateral-line nerve occurs 
in each body segment. The fourth and fifth are the largest and come close to- 
gether as they pass anteriorly around the base of the pectoral fin. The rest form 
an interlocking network on the ventrolateral side where they innervate free 
neuromasts. 

Pattern of RLA in Osmeridae 

In Hypomesus pretiosus there is the same type of nerve as in Percopsis that 
leaves the truncus hyomandibularis and courses over the wall of the otic bulla, 
that is crossed by the ninth cranial nerve, that forms ganglia along its trunk, 
and that courses alongside the ventrolateral side of the vertebral column as far 
as the caudal fin. 

In osmerids, however, this nerve is the sympathetic trunk and ganglia. The 
presence of RLA in it cannot yet be demonstrated. From leaving the truncus 
hyomandibularis up to halfway across the otic bulla, the sympathetic consists 
of 3 separated parts. Proximally, two of these parts pass anteriorh^ inside the 



238 



CALIFORNIA ACADEMY OF SCIENCES LProc. 4th Ser. 



EXTERNAL FIN RAY NERVE 




Figure 12. Pelvic fin of Physiculus talarae (Moridae) showing fin-ray nerves coursing 
external to fin rays. Sihler preparation. 



cranium. They do not end in the geniculate ganglion. The third could not be 
followed. It is hypothesized that in some osmerid it may be found to contain 
RLA fibers. This prediction is based on the fact that osmerids and percopsids 
have both the ramus canalis lateralis system of cranial nerves as well as the pec- 
toral accessory ventral lateralis nerves and the similar successive segmental ven- 
tral lateralis nerves for all body segments. Osmerids may be basically much like 
Percopsis and may have given rise to them. 

THE RAMUS CANALIS LATERALIS SYSTEM 

The ramus canalis lateralis nerves first named by Ray (1950) for the lantern 
fish, Lampanyctus leucopsarus, are here recognized as evidently a secondary 
system of lateralis innervation and perception that is found in numerous lower 
groups of fishes, the primary system being the neuromasts located in the cephalic 
lateral -line canals. 

The ramus canalis lateralis system consists of several facialis and occasionally 
of one or more vagal lateral-line branches that usually course lengthwise an- 
teriorly in the membranous roof of the cephalic lateral-line canals. Several of 
these canal branches radiate from a short trunk coming off the truncus hyo- 



Vol. XXXVIII] 



FREIHOFER: NERVE PATTERNS 



239 



NTERNAL FIN RAY NERVE ; 




VEIN RAY 



:r.^\ 




FiGXTRE 13. Caudal fin rays of Aphredoderus sayanus showing fin-ray nerves coursing 
internally down the centers of the hollow fin rays. Sihler preparation. 



mandibularis close to the articulation between the opercle and hyomandibular 
bones. Although the system of branches looks complex (fig. 16), the presence 
of the system is simple to detect. If the skin is removed in the upper cheek 
region posterior to the eye, the most conspicuous branch will be seen. It comes 
up through cheek muscle close to the preopercle and runs forward beneath skin 
and into the cavity of the middle infraorbital bones and out along the overlying 
membrane of the anterior infraorbitals. This branch has been called the ramus 
buccalis accessorius. Branches of the canalis lateralis nerves innervate naked 
lateral-line organs (neuromasts) lying in the membrane roofing the canals and 
also innervate the same kind of organs lying in skin adjacent to the canals if 
such organs are present. The ramus canalis nerves do not supply the neuromasts 
(seismosensorial organs) lying in the floor of the head canals. The branches of 
the ramus canalis system appear to have great taxonomic importance. Patterns 
vary with different groups of fishes or are the same or similar for other groups. 
As far as presently known, these nerves are found only in certain fish groups of 
lower taxonomic placement; that is, not in the Perciformes or higher orders, or 
if they are, then the classification of such fishes should be questioned. 



240 



CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 



N RAY 




EXTERNAL FIN RAY NERVE 



Figure 14. Pelvic fin of Brotula clarkae showing fin-ray nerves coursing external to 
fin rays. Sihler preparation. 



The descriptions v^^hich follow are brief and preliminary. The ramus canalis 
lateralis nerves will be given extensive treatment in a later publication. Percopsis 
omiscomaycus is used as the basic reference form to which other forms are com- 
pared. Descriptions are reduced to cover 7 rami. The basic branches and their 
abbreviations are as follows: 

r.c. la = supraorbital branch. 

r.c. lb = temporal branch of r.c. la. 

r.c. 2 = anterior infraorbital branch plus dorsoanterior dentary and rictus 
branches. 

r.c. 3 = posteroventral dentary branch plus rictus branches. 

r.c. 4a = preopercular branch. 

r.c. 4b = medial preopercular-mandibular ridge prolongation of r.c. 4a. 

r.c. S = supratemporal branch. 

r.c. 6 = posterior infraorbital branch. 

r.c. 7 = branch from r.c. 4 arching dorsoposteriorly from operculum to an- 
terior end of dorsal fin. 



Vol. XXXVIII] FREIHOFER: NERVE PATTERNS 241 

Some patches of naked lateralis organs are supplied by nerve fibers that 
course indistinguishably with the supraorbital and infraorbital trunks. These 
fibers are not considered further in the present report although they probably 
are part of the same system of nerve fibers as are the ramus canalis lateralis 
nerves. 

Ramus Canalis Lateralis System in Percopsidae 

The branches of the ramus canalis lateralis system for Percopsis omisco- 
maycus are shown in figure 15. Branches r.c. 1 through r.c. 5 are present and 
well-developed. No group other than the percopsiform fishes yet examined has 
branch r.c. 4b, the extension of r.c. 4a into the medial preopercular-mandibular 
ridge. The approximate distribution of the naked lateralis organs suppHed by 
the ramus canalis lateralis system of the head is also depicted in figure 15. The 
branches are closely similar for the Aphredoderidae. 

Ramus Canalis Lateralis System in Myctophidae 

The ramus canalis lateralis nerves have been reported only in Lampanyctus 
(see Ray, 1950). A more complete description of them is given in figure 16 of 
the present work. The skin in lantern fishes is often damaged in capture. Com- 
plete descriptions of the nerves and the distribution of the organs they innervate 
therefore await specimens with nearly intact skin and organs. 

Branch r.c. la is basically the same in both Percopsis and Lampanyctus 
except that the temporal branch, r.c. lb, is absent in the latter. Branch r.c. 2 
is essentially the same also if allowance is made for the very large mouth in 
Lampanyctus . Branch r.c. 3 is basically like that in Percopsis, but comes off 
the truncus hyomandibularis further down in Lampanyctus. Likewise for r.c. 4, 
except r.c. 4 is not prolonged into the medial preopercular-mandibular ridge, as 
is branch r.c. 4b. Branch r.c. 5 innervates the temporal and posttemporal canal 
membrane in Lampanyctus, a difference from what r.c. 5 innervates in Percopsis, 
where it supplies a patch of organs medial to the small supratemporal canal bone. 

Ramus Canalis Lateralis System in Apogonidae 

Free or naked neuromasts, all over the head in Apogon, are randomly but 
densely distributed on the snout, becoming progressively arranged in definite 
rows posteriorly. 

Branch r.c. 1 seems to be basically the same as in Percopsis with both sim- 
ilar "a" and "b" branches (fig. 17). Branches r.c. 2 and r.c. 3 arise together as 
a main unit from the truncus hyomandibularis, whereas in Percopsis these 
branches arise separately. This origin of r.c. 2 and 3 as a common trunk would 
seem to explain the fact that r.c. 2 has no visible branch to the lower jaw as in 
Percopsis. That is, since r.c. 2 and r.c. 3 have a common trunk going back to 
the truncus hyomandibularis, it would seem likely that the part of r.c. 2 that 



242 



CALIFORNIA ACADEMY OF SCIENCES 



LProc. 4th Ser. 




Ph 



a, ^ 



^ fl 



Vol. XXXVIII] 



FREIHOFER: XERVE PATTERNS 



243 




244 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

goes to the lower jaw as a separate branch in Percopsis would remain bound in 
Apogon with r.c. 3 which also goes to the lower jaw. Branches which go to the 
part of the preopercular canal that lies in the angle of this bone detach in Per- 
copsis from the truncus hyomandibularis, but they detach from r.c. 3 in Apogon. 
Branch r.c. 4 is as it is in Lampanyctus; that is, without the extension of r.c. 4b. 
Branch r.c. 5 is similar to that of Percopsis. Branch r.c. 6 is apparently not rep- 
resented in Lampanyctus or Percopsis. The branches in Apogon are in some ways 
more like those in Lampanyctus and in others more like those in Percopsis. Syna- 
grops bella has a large branch r.c. 3 and branch r.c. 5 but apparently no r.c. 1 or 
r.c. 2 or r.c. 4. Epigonus robustus has a smaller r.c. 3 than has Synagrops but 
apparently none of the other independent rami. 

Ramus Canalis Lateralis System in Neoscopelidae 

On Scopelengys tristis (fig. 19) free neuromasts are variously arranged in 
short and long rows on all parts of the head. The pattern of the branches of the 
ramus canalis lateralis system in Scopelengys is closest to that of Lampanyctus 
but with differences. Closely similar are branches r.c. la and r.c. 2 in Percopsis, 
Lampanyctus and Scopelengys. Branch r.c. 3 has a distinctive branch, r.c. 3a, 
that goes to the membrane of the infraorbital canal lying above the posterior end 
of the maxillary ramus. The same branch is present in Lampanyctus. Branch 
r.c. 3b detaches from the truncus hyomandibularis further distally. Scopelengys, 
Lampanyctus, and Apogon all agree in having all of branch r.c. 3 come off the 
truncus hyomandibularis as one branch, whereas in Percopsis several smaller 
branches supplying the angle of the preopercular canal detach from different 
points of the truncus hyomandibularis. In Scopelengys branch r.c. lb detaches 
independently from r.c. la, and branch r.c. 3c, not found in the others, detaches 
from the common trunk of the ramus opercularis superficiaHs facialis. No 
branches associated with the ramus supratemporalis vagi were observed. Neo- 
scopelus differs interestingly from Scopelengys in having r.c. 7 which extends in 
an arch from near the opercular articulation dorsoposteriorly back to the anterior 
end of the dorsal fin, decreasing in size and ending there. 

Ramus Canalis Lateralis System in Melamphaidae 

Only one specimen of Melamphacs species prepared by the Sihler technique 
was available. Free neuromasts are extremely abundant on the head, especially 
in the membranous roofs of the cephalic canals as in Lampanyctus. 

The ramus canalis lateralis nerves are thin. Branches r.c. la, r.c. lb, and 
r.c. 2 are present. No independent r.c. 3 was observed. Branch r.c. 3 runs with 
the ramus mandibularis facialis to the lower jaw as part of the latter. Branch r.c. 
4 extends at least halfway down the preopercular canal. No branch r.c. 5 was 
observed . 



Vol. XXXVIII] 



FREIHOFER: NERVE PATTERNS 



245 










[3^ 



246 CALIFORNIA ACADEMY OF SCIENCES LProc. 4th Ser. 

Ramus Canalis Lateralis System in Meelucciidae 

In Merluccius gayi branch r.c. 1 does not come off the truncus hyomandi- 
biilaris as a separate branch at the same point that the ramus opercularis super- 
ficialis facialis does in Pcrcopsis. Instead it detaches from the truncus supra- 
orbitalis together with the branch that supplies the third neuromast from the pos- 
terior end of the supraorbital canal. From there on it runs separately in the 
membranous roof of the canal to its anterior end. 

Branch r.c. 2 detaches from the truncus hyomandibularis high up on the 
cheek. It has two divisions: one to the membranous roof of the infraorbital canal 
all the way to its anterior end and one to the membranous roof of the mandib- 
ular canal up to its anterior end. 

Branch r.c. 3 runs in common with, then separates from, r.c. 2, and goes to 
the lower jaw, giving off en route a branch to the angle of the preopercular canal 
roof and to a long horizontal row of organs located midway up the cheek. The 
rest of r.c. 3 innervates the posterior end of the mandibular-articular part of the 
canal membrane and extends along the membranous roof of the canal to its an- 
terior end. 

Branch r.c. 4 detaches from the ramus opercularis superficialis facialis and 
runs ventrally down the membranous roof of the preopercular canal to the angle 
of this canal. 

Ramus Canalis Lateralis System in Brotulidae 

In Dicrolcnc intronigra one large branch of the ramus canalis lateralis system 
is present. The first branch detached from it is small and passes to the infra- 
orbital canal below the eye and ends there. The remainder passes to the lower 
jaw, giving off branches en route supplying large free neuromasts lying near the 
angle and horizontal arm of the preopercle, the rest supplying free neuromasts 
on the membranous roof of the mandibular canal extending almost to its an- 
terior end. 

Ramus Canalis Lateralis System in Osmeridae 

The distribution of free neuromasts for Hypomesus pretiosus (fig. 18) is 
generally over all of the head even on the maxillary and supramaxillary and on 
exposed branchiostegal rays as well as on the membranes of all the fins and on 
the leading edges of the pectoral, pelvic, dorsal, and anal fins. 

Branch r.c. la and r.c. lb are both large and long. Branch r.c. lb arches pos- 
teriorly all the way to the segment of the lateral-line canal attached to the supra- 
cleithrum. Branch r.c. la extends to the nasal canal. 

Branch r.c. 2 runs along the infraorbital canal to its anterior end and de- 
taches a branch which drops vertically down across the cheek to the canal of 
the horizontal arm of the preopercle. There is no independent branch of r.c. 3 



Vol. XXXVIII] 



FREIHOFER: NERVE PATTERNS 



247 




248 



CALIFORNIA ACADEMY OF SCIENCES 



LProc. 4th Ser. 




Vol. XXXVIII] FREIHOFER: XERVE PATTERNS 249 

to the lower jaw. Branches innervating free neuromasts of the preopercular- 
mandibular canal detach from the ramus mandibularis facialis to the lower jaw. 
Practically the same pattern is present in Spirinchus thaleichthys and Thal- 
eichthys pacijiciis as seen from dissections on alcoholic specimens; the same 
holds for Bathylagus alascaniis. 

Ramus Canalis Lateralis System in Gobiidae 

In Bathygobius lineatus as in all gobioids there are various rows of free 
neuromasts associated with most of the cephalic head canals. These are inner- 
vated by several branches of what evidently is the ramus canalis lateralis system. 
Branch r.c. 1 is short, extending only from where the superficial opercular facial 
ramus passes onto the operculum and on up the dorsal end of the preopercle, 
ending at the posterior end of the temporal canal. The shortness of r.c. 1 re- 
flects the fact that there are no free neuromasts for the supraorbital canal except 
at its anterior end. 

Branch r.c. 2 is the main independent branch of the ramus canalis system 
that is present. At the point where the superficial opercular facial ramus de- 
taches from the truncus hyomandibularis, the first branch of r.c. 2 also detaches 
and innervates a long line of naked neuromasts extending from just posterior 
of the eye to a little below the middle of the eye. A row lies some distance dorsal 
to this row and another somewhat below this dorsal row that together form a 
somewhat continuous row innervated by several branches from the truncus infra- 
orbitalis. The main part of r.c. 2 continues forward beneath the skin innervating 
the ventralmost row of free neuromasts lying below the eye and in the area of 
the infraorbital canal. The row ends some distance anterior and dorsal to the 
rictus of the jaw. The main branch of r.c. 2 next detaches 2 branches that supply 
a row of free neuromasts that lie along the lateral (or dorsal) edge of the pre- 
opercular-mandibular canal, beginning with the angle of the preopercle and con- 
tinuing forward as far as the angular bone. A row of free neuromasts lying along 
the medial edge of the same length of canal is supplied by branches from the 
ramus mandibularis facialis. Shortly before the rictus of the mouth, branch r.c. 
2 divides. The smaller part innervates the remainder of the row of free neuro- 
masts already mentioned that ends shortly beyond and above the rictus. The 
larger part of r.c. 2 passes ventrally around the rictus of the mouth and onto the 
lower jaw where it extends anteriorly to near the symphysis innervating a long 
row of free neuromasts lying on the lateral or dorsal edge of the mandibular canal. 
The row of larger organs extending along the medial edge of the mandibular 
canal is supplied by branches from the ramus mandibularis facialis. 

Branch r.c. 4, or what appears comparable to it, has two main parts. One con- 
tinues posteriorly about on the level of the horizontal rib of thickened reinforce- 
ment bone of the opercle and supplies a row of 24 organs on the posterior third 
of the opercle. The larger part of branch r.c. 4 passes down the posterior edge 



250 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

of the preopercular canal or close to it, out onto the ventral end of the preopercle 
and supplies a long vertical row of free neuromasts. 

Branch r.c. 5 of the ramus supratemporalis vagi innervates free organs along 
the posttemporal and supratemporal canals and another branch passes forward 
supplying the temporal canal. 

Ramus Canalis Lateralis System in Cyprinidae 

Manigk (1934, fig. 1) reports for Phoxinus laevis a nerve which he calls the 
ramus buccalis accessorius that apparently belongs to the ramus canalis lateralis 
system but which has important differences from the patterns seen in the Per- 
copsidae, Myctophidae, Osmeridae, Apogonidae, and Gobiidae. The pattern in 
the Cyprinidae differs in that ( 1 ) the rami do not course in the membranous 
roof of the cephalic lateral-line canals; (2) there is only one distinct branch, r.c. 
2; and that (3) branch r.c. 2 does not have a ramus going to the lower jaw as 
occurs in all other families except the most generalized, the Osmeridae. 

Ramus Canalis Lateralis System in Other Families 

Other families among those examined having the ramus canalis lateralis 
system are, in the Salmoni formes, the Esocidae, Umbridae, Gonostomatidae, 
Sternoptychidae, Chauliodontidae, Alepocephalidae, and Chloropththalmidae ; 
in the Beryciformes, the Trachichthyidae, Berycidae, Polymyxiidae, and Holocen- 
tridae. The Umbridae have r.c. 1, r.c. 2, r.c. 4, and r.c. 5. The pattern in Gono- 
stoma elongatum is most like that in the Neoscopelidae and Myctophidae. 

The beryciform families apparently have the ramus canalis lateralis pattern 
of Melamphaes or it is reduced. 

Found not to have the system of nerves were the Amiidae; Elopidae; Clupe- 
idae (Clupea pallasii); Salmonidae (Salmo, Oncorhynchus, Coregonus); Gala- 
xiidae (Galaxias); Synodontidae ; Atherinidae (Menidia); Plecoglossidae ; Ar- 
gentinidae. 

EXTERNAL-INTERNAL FIN-RAY INNERVATION 

So far as is known, most of the paracanthopterygian fishes have the fin-ray 
nerves coursing external to the fin rays, not internal in the space between the two 
halves of each fin ray as is true for percoid fishes and apparently for most other 
teleosts. 

In the survey made thus far, the external position of the fin-ray nerves has 
been found for the codfishes, brotulids, ophidiids, zoarcids, gobiesocids, batra- 
choids, and ogcocephalids that have been examined. When external, the nerves 
course in contact with the surface of the fin ray or almost in contact. The fin- 
ray nerves are internal for the percopsiform fishes. 

In Gadopsis marnioratus, a fresh-water percoid fish of Australia, in which 
the pelvic fins are long, narrow, and of 2 rays, similar to pelvic rays in bro- 



Vol. XXXVIII] FREIHOFER: SERVE PATTERNS 251 

tulids, the huge combined spinal and RLA nerves to them course down the cen- 
ters of the rays with some fibers, apparently out of physical limitation of space, 
coursing outside the fin rays beside the longitudinal split that exists down the 
two halves of each ray. The nerves of ophidiid and brotulid pelvic rays are huge 
also, but the nerves course entirely external to the pelvic rays. The fin ray inner- 
vation in Gadopsis is not ophidioid. 

The fin-ray nerves are external also in the Liparidae and Cottidae {Scor- 
paenichthys) . They are external also in the stichaeids (Epigeichthys). Thus 
the condition of the fin-ray nerves being external to fin rays is not unique to 
the Paracanthopterygii, but it is apparently a specialization within the fishes 
of this group. A survey for this interesting condition is being conducted with 
the preparation of specimens of many families by the Sihler technique in addi- 
tion to some serially sectioned. 

DISCUSSION 

New facts presented in preceding sections have bearings on the taxonomic 
position and phylogenetic considerations of numerous kinds of fishes. Only some 
of the problems can be discussed here. 

The new facts concern 4 nerve patterns: (1) the ramus lateralis accessorius 
(RLA); (2) the pectoral accessory ventral lateralis branch and succeeding 
segmental branches; (3) the ramus canahs laterahs system of nerves; (4) the 
external or internal innervation of fin rays. 

ORIGIN OF BATRACHOIDID FISHES 

The presence of greatly similar patterns in the first two of these nerve com- 
plexes in batrachoidid and percopsid fishes indicates that the batrachoidid fishes 
have most probably been derived from percopsid ancestors. 

The patterns of RLA in Percopsis and Porichthys are similar in three impor- 
tant ways in that in each genus ( 1 ) RLA leaves the cranium together with the 
truncus hyomandibularis; (2) RLA leaves the truncus hyomandibularis bound 
up with the sympathetic trunk; (3) RLA courses over the wall of the otic bulla. 
Porichthys differs from Percopsis in that RLA departs from the sympathetic 
trunk shortly after the two leave the truncus hyomandibularis bound together. 
RLA courses across the otic bulla slightly diverging from the sympathetic. Por- 
ichthys differs also in its course beyond the cleithrum. In Porichthys RLA breaks 
up into the dorsal, anal, and pectoral-pelvic branches near the cleithrum. these 
branches going independently to their fins, whereas in Percopsis RLA continues 
along the vertebral column to the caudal fin bound to the sympathetic trunk. 
The pattern of RLA in Porichthys appears to be an intermediate stage or an off- 
shoot development between the Percopsis pattern and a gadoid or ophidioid 
pattern. 

Porichthys has a pectoral-pelvic accessory ventral lateralis nerve as has 



252 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

Percopsis and also has several of the succeeding segmental ventral lateralis 
branches enlarged which supply sections of the ventral lateral lines. Percopsis 
lacks ventral lateral lines, having instead only scattered, free, lateralis organs 
mostly in the pectoral-pelvic area. Porichthys does not have independent 
branches of the ramus canalis lateralis system. It has the external rather than 
internal fin-ray innervation which Percopsis has, but, so far as is known, all 
paracanthopterygian fishes have the external fin-ray innervation pattern except 
percopsi forms. 

The pattern of RLA is the most distinctive feature of similarity between 
batrachoidid and percopsiform fishes. It more than any other character points 
to a percopsiform origin for batrachoidids. Possession of accessory pectoral- 
pelvic and ventral segmental laterahs branches is almost as significant and also 
helps tie batrachoidids in turn to other lower taxonomically placed fishes having 
this same nerve development. 

ORIGIN AND RELATIONSHIPS OF GADOID AND 
OPHIDIOID FISHES 

There are several patterns of RLA in ophidioids indicating evolutionary di- 
vergence within the group. The pattern in the brotulid Ogilbia (fig. 2) is most 
similar to that in a gadoid such as Physiculus (fig. 4) being alike in all im- 
portant points, especially in having two or more anal fin branches detaching from 
the dorsal fin branch posterior to the origin of the dorsal fin. The pattern of 
Lepophidium and of numerous other ophidioids is like that of Ogilbia and Physi- 
culus except that the branch for the dorsal fin and that for the anal fin exit 
from the cranium through separate foramina some distance apart in the parietal 
bone. The pattern in Brotula is distinctive for an ophidioid or gadoid in that 
the pectoral-pelvic branch leaves the cranium together with the truncus hyoman- 
dibularis and then courses posteriorly, passing medial to the supracleithrum. 
Such a cranial exit is an important similarity to the pattern in Percopsis and 
Porichthys. 

The presence in the brotulids Dicrolene, Monomitopus, Monomerepus, and 
Porogadus, of the accessory pectoral-pelvic ventral segmental lateralis nerve and 
several succeeding similar branches greatly similar to such nerves in Percopsis 
and Porichthys, is evidence also for a taxonomic and phylogenetic connection 
between these fishes. So is the presence in both the gadoid Merluccius and the 
brotulid Dicrolene of significant branches of the ramus canalis lateralis nerves, 
although these nerves are somewhat different in parts of their patterns from 
that in percopsiform fishes. The pattern of the fin-ray nerves coursing external 
to the fin rays in gadoids and ophidioids (as well as in all paracanthopterygian 
fishes except percopsiforms) is a further interesting and significant similarity 
when it is contrasted with the internal fin-ray innervation of percoid fishes (20 
families examined) and numerous other large groups of fishes such as atherinids. 



Vol. XXXVIII] FREIHOFER: XERVE PATTERNS 253 

cyprinodonts, salmonids and beryciforms. Other important similarities between 
gadoid and ophidioid fishes are (1) the large opisthotic, occupying part of the 
otic bulla wall, separating the prootic and exoccipital bones, and having the 
ninth cranial nerve exiting through it (McAllister, 1968); (2) presence of 
levator maxillae superioris muscle (Greenwood et al., 1966) ; (3) a "percopsiform 
projection" often present on middle branchiostegal rays (McAllister, 1968). 

RELATIONSHIPS OF ZOARCID FISHES 

The Zoarcidae is still best placed together with the gadoids and ophidioids 
(Rosen, 1962; Freihofer, 1963; Greenwood et al, 1966). The zoarcids have a 
pattern of RLA that is most like that of these fishes, especially like that of the 
ophidioids. Some zoarcids appear to have a remainder of the levator maxillae 
superioris muscle (Greenwood et al., 1966). Lycodapus has what must be a 
good example of this muscle. Zoarcids have an external fin-ray innervation 
pattern. 

Examination for RLA in Lycodapus has not yet been possible. Gosline 
(1968) has questioned the zoarcid affinities of Lycodapus and also tried to show 
that ophidioids and gadoids are not basically similar but that ophidioids are per- 
coid derivatives. On the basis of patterns of RLA, the zoarcids most probably 
do not have a percoid ancestry as Gosline maintains, nor do they on the basis of 
ventral segmental lateral-line nerves present in zoarcids but not in percoids as 
far as is known. 

RELATIONSHIPS AND CLASSIFICATION OF GOBIOID FISHES 

Gobioid fishes show external similarities to percopsid and amblyopsid fishes. 
Gobioids and percopsids have weak spinous dorsal fins, rows of free lateralis 
organs on the head, and similar body form. McAllister (1968) lists many partial 
similarities between gobioids and percopsiforms and almost placed them in a re- 
lationship to the percopsiforms in his classification of teleostome fishes but could 
not quite do it. The problem has been that many of the similarities of gobioids 
to percopsiforms are halfway similarities, none being great enough to be con- 
vincing. Two important similarities of the nerves of gobioids and percopsiforms 
are now known. These are the presence of part of the ramus canalis lateralis 
system and of an accessory pectoral ventral lateralis branch. The only other 
spiny-rayed fishes so far known to have both of these nerve features are the 
Percopsiformes. This fact, plus all the other varying degrees of similarity listed 
by McAllister, suggests that the affinity of the gobies with the Percopsiformes 
is as he suspected. Examination of the opisthotic of Eleotris juscus shows that 
it is large, separates the prootic and exoccipital, and has the ninth nerve coming 
out near its center. These opisthotic features in Eleotris are also those of gadoid 
and ophidioid fi.shes. The intermediateness of the condition of the opisthotic in 
Eleotris is suggested by the fact that the foramen for the ninth nerve on one side 



254 C ALIFORM A ACADEMY OF SCIENCES IProc. 4th Ser. 

of Eleotris jusciis was connected by a long thin opening to the edge of this bone; 
on the other side the indentation did not reach the ninth nerve foramen. 

Bathygobius Uneatus shows a great reduction in lateralis organs located in 
the cephalic canals, only a few being present in the region of the upper preoper- 
cular and temporal canals. This may be a feature of all gobioids. The canals 
themselves are mostly open troughs. These features of the head canals of gobi- 
oids should have interesting ecological, behavioral, and physiological significance. 
Their usually small size, many lines of free lateralis organs on the head with as- 
sociated near loss of canal organs, and relatively sedentary habits give credence 
to a depiction of gobies as microhabitat fishes. 

With new nerve evidence and the many features of similarity of gobies to 
percopsiform fishes as given by McAllister, it seems justified to make the gobioids 
an order, the Gobiiformes, placed in the superorder Paracanthopterygii with 
closest relationships to the Percopsiformes and with other characters as given by 
Regan (1911) for his suborder Gobioidea. 

RELATIONSHIPS AND CLASSIFICATION OF APOGONIDAE 

Nerves identified as the ramus canalis lateralis system in Apogon (fig. 17) 
are strikingly similar to those of percopsiforms (fig. 15), myctophoids (figs. 16 
and 19), stomiatoids, and osmerids (fig. 18). The ramus canalis lateralis system 
is recognized in an apparently reduced form in gadoids, ophidioids, beryciforms, 
and gobioids. If this system of nerves in Apogon is homologous with that in the 
other groups listed, it would be an important preperciform feature found even in 
salmoniforms but unknown in any perciform group except the gobioid fishes. 
That it is homologous in all these groups needs to be rigorously shown but 
judging from the similarity of the patterns illustrated in this paper I think that it 
is broadly homologous in these groups. It is one of the most complex and distinct 
system of nerves on the head. Its absence as far as known in the perciforms is 
explained as loss through reduction in preperciform ancestors. The system is 
apparently considerably reduced in beryciforms or absent in some. The great 
development and similarity of the canalis lateralis system in apogonids to that 
in Percopsis, Scopelengys, and melamphaeids is strong evidence for questioning 
the perciform character of apogonids. 

There is also evidence of an apogonid-salmopercoid relationship from the 
otoliths. Frost's systematic statements, based as they were only on characters 
of the otoliths and made from study of otoliths of only certain species, have 
often been confusing. An example is in the salmopercoid fishes. Frost (1925) 
says that their otoliths "closely resemble (in some respects) those of Ophichthys 
gomesii (Apodes). — On the other hand there is a strong resemblance to those 
of the percoid genus Apogon which differs from the remainder of the percoids 
in the sulcus." We still do not have any other characters pointing to a system- 
atic relationship between apodal and salmopercoid fishes but we do have an 



Vol. XXXVIII] FREIHOFER: .\ERVE PATTERS S 255 

important similarity in nerves identified in both salmopercoid and apogonid 
fishes as belonging to the ramus canalis lateralis system. Frost (1927) also 
states that "the sagitta of Apogon melanotaenia resembles that of Acerina (Per- 
cidae) in the sulcus" and that the otolith of Cepola rubescens (Cepolidae) re- 
sembles closely that of Apogon. The systematic significance of these similarities 
remains doubtful. Without doubt, however, there is a great need for atlases of 
fish otoliths published family by family until all families are covered. 

A reasonable alternative hypothesis suggested here to that of Regan (1913) 
and of Greenwood ( 1966) for the origin of the Perciformes is that the percopsi- 
forms gave rise to lines leading to both the Paracanthopterygii and the Acan- 
thopterygii. In this hypothesis apogonids could lie on one of the lines tending 
towards the percoid expression. It is suggested that the Apogonidae be tried in 
a preperciform position in an experimental classification as a suborder of bery- 
ciform fishes allied to percopsiform fishes as all beryciforms may be so allied. 

ORIGIN OF PERCOPSIFORM FISHES 

Several important facts point to an origin of the percopsiform fishes from 
the salmoniforms, in particular, from the Osmeridae. The hypothesis in Green- 
wood et al. (1966)^ leaves room for other possibilities. 

The fact most weakening to the Greenwood hypothesis is that the nerve 
evidence in the descriptive section of the present paper indicates a percopsiform 
origin for the batrachoidids, gadoids, and ophidioids. In the Greenwood hypoth- 
esis, these fishes, including the percopsiforms, are each independently derived 
from a neoscopelid-like ancestor (e.g., Sardinioides), which in turn is derived 
from myctophoids. 

In the percopsiform hypothesis, the myctophid and neoscopelid-like fishes 
evolved from a common ancestry with percopsiform fishes. Weitzman (1968) 
has shown osteologically that the origin of stomiatoid fishes evidently is in the 
Osmeridae. With present knov.'ledge of the nerves, an osmerid ancestry is also 
hypothesized for the percopsiform fishes. The osmerids have a more generalized 
pattern of the ramus canalis lateralis system and also of the pectoral accessory 
ventral lateralis nerves and succeeding ventral segmental lateralis branches. 
Neither of these systems of nerves is present in the Salmonidae or Plecoglossidae 
as far as known. The Osmeridae are completely generalized in their freshwater- 
saltwater tolerance, a fact that a basal ancestral group for such a radiation should 
be expected to have. An observation of osmerid otoliths by Frost (1925) is of 
greater interest now. Frost said "In osmerus eperlanus .... there is a distinct 
advance in this form (of the otolith) towards those observable in later groups, 
notably the berycoids." 



1 The important paper of Rosen and Patterson, The structure and relationships of the paracant hopterygian 
fishes. Bulletin of the .\merican Museum of Natural History, vol. 141, art. 3, pp. 357^74, came too late 
(July, 1969) to be used. 



2S6 CALIFORNIA ACADEMY OF SCIENCES IProc. 4th Ser. 

The ramifications of the percopsiform hypothesis are shown in figure 20. 

No RLA is yet known for the osmerids nor for the suborder Myctophoidei 
nor is the pectoral accessory lateralis or succeeding segmental branches present 
in the myctophoids, Lampancytus and Scopelengys. The ramus canalis lateralis 
system apparently is not present in Aulopus, Harpodon, and Synodus. 

In the words of its original describer, Louis Agassiz, no better candidate 
exists among known teleosts for being "a true intermediate type between Per- 
coids and Salmonidae than my Percopsis" .... as it "shows peculiarities which 
occur simultaneously in the fossil fishes of the chalk epoch, which however soon 
diverge into distinct families in the tertiary period, never to be combined again" 
(Agassiz, 1850). 

Aphredoderus also is important in this picture but must be taken up in a 
later paper. 

SUMMARY OF SYSTEMATIC RESULTS 

Various combinations of four general features of the nerves plus some specific 
nerve characters were used to arrive at the systematic interpretations discussed 
above and summarized below. The nerve complexes are ( 1 ) the ramus lateralis 
accessorius (RLA or recurrent facial nerve), (2) the ramus canalis lateralis 
facialis system, (3) the ventral segmental lateral-line nerves of the trunk, and 
(4) the external or internal pattern of fin-ray innervation. The systematic re- 
sults are as follows: 

(a) Ophidioid relationships lie with gadoid fishes as advocated by Rosen 
and his coworkers (1962, 1966, 1969) and Freihofer (1963), not with blenniid- 
like perciforms. More complete descriptions in the present paper of RLA in 
these fishes show patterns of RLA to be more important systematically and more 
alike in these groups than was first thought. Possession of the ramus canalis 
lateralis facialis system and of ventral segmental lateral-line nerves in gadoids 
and ophidioids add strong support to the gadoid placement of the ophidioids. 
Both groups have the external pattern of fin-ray innervation. 

(b) Specific relationship of ophidioids with percopsids is shown by cranial 
exit of RLA through the facial foramen beside the truncus hyomandibularis in 
Brotula and Percopsis. Similar cranial exit is known for Anguilla, the cobitid 
Nemacheilus, and the goatfish Parupencus, but it is considered convergent in 
these groups. Doubt about the percoid affinities of the peculiar goatfishes is 
raised. Presence also of both the ramus canalis lateralis facialis system and of 
ventral segmental lateral-line nerves in the brotulid Dicrolene, and of the latter 
system and probably the former system also in the brotulids Monomitopus, 
Monomerepus, and Porogadus, are important nerve features found also in stomi- 
atoid, myctophoid (only the ramus canalis lateralis system), and osmerid fishes 
but of neither system in salmonid fishes as far as known. 



Vol. XXXVIII] 



FREIHOFER: NERVE PATTERNS 

PERCl FORMS 



257 



EERYCIFORMS 



PARACANTHOPTERYGIANS 




ATHERINOMORPHS 



PERCOPSIFORMS 



MYCT0PH0ID5 



STCMIATOIDS 




OSyERIDS 
Figure 20. Scheme of possible relationships of percopslform fishes. Two alternative 
hypotheses of relationship to perciforms are shown, one direct and one by way of the bery- 
ciforms. 

(c) Gadoids have an affinity with percopsids as shown by the presence in 
Merluccins of both the ramus canaHs laterahs faciahs system and of ventral, 
segmental lateral-line nerves on the trunk, the latter more generalized in Mer- 
luccius than in Percopsis. 



258 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

(d) Batrachoidids have an affinity with percopsids. Porkhthys, Percopsis, 
and Amblyopsids are the only fishes known to have RLA leave the cranial cavity 
through facial foramen and bound with the trunk of the sympathetic nervous 
system. Porkhthys possesses ventral segmental lateral-line nerves but in a more 
specialized condition than in Percopsis. 

(e) Amblyopsids have a closer affinity with percopsids than with Aphredo- 
derus. Pattern of RLA in Percopsis is found also in Chologaster and Porichthys. 
It is judged more primitive than the pattern in Aphredoderus, a pattern much 
different from that in Percopsis and more like that found in gadoids, ophidioids 
(except for the percopsid-like cranial exit in Brotula), and higher fishes. 

(f) A percopsid-like ancestry is hypothesized for batrachoidid, gadoid, and 
ophidioid fishes based on evidence from the nerves. 

(g) RLA in Percopsis, Porichthys, and Physicnlus are described for the first 
time. 

(h) Pattern of RLA in Macruridae is more like that in Moridae than in 
Gadidae. 

(i) Pattern of RLA in Ogilbia and Dinematichthys resembles that in Mori- 
dae and Gadidae more than does RLA pattern in other brotulid and ophidiid 
fishes. Ogilbia and Dinematichthys and related genera merit familial rank. 

(j) A well developed ramus canalis lateralis facialis system is present in os- 
merids, stomiatoids, myctophoids, percopsiforms, gadoids, ophidioids, and batra- 
choidids; reduced patterns are present in berycomorphs; the system is absent 
in advanced paracanthopterygians and in perciforms, except in gobioids and 
apogonids. 

(k) Ventral segmental lateral-line nerves are present in osmerids, stomia- 
toids, percopsiforms, gadoids, batrachoidids, ophidiids, zoarcids, and gobioids. 

(1) Presence of ventral segmental lateral-line nerves in zoarcids is strong 
evidence for continued placement of zoarcids close to gadoids and ophidioids. 
Zoarcids also have the external pattern of fin-ray innervation. 

(m) External pattern of fin-ray innervation is found only in certain groups: 
all paracanthopterygian fishes, except percopsiforms and gobiiforms (placed in 
Paracanthopterygii in present paper) ; scorpaenids, cottids, liparids, stichaeoids. 
The survey is incomplete. Apparently most fishes have internal pattern of fin- 
ray innervation. Presence of the external pattern in nonparacanthopterygian 
fishes is considered convergent. 

(n) Osmerids have the ramus canalis lateralis facialis system and ventral 
segmental lateral-line system in the most generalized state but have no RLA. 
Stomiatoids have the first two systems and lack RLA as far as known. Sal- 
monids, aulopids, and synodontids lack all three systems so far as known. It is 
hypothesized that percopsids have an osmerid ancestry. 

(o) Apogonids may be quasi-percoid. They possess a well developed non- 



Vol. XXXVIII] FREIHOFER: NERVE PATTERNS 259 

perciform ramus canalis lateralis facialis system. It is proposed that they be 
placed with the beryciforms as a percopsid offshoot. 

(p) Nerves identified as belonging to the ramus canalis lateralis system have 
been found in gobioid fishes. One anterior nerve identified as belonging to the 
ventral segmental lateral-line system known elsewhere only from preacanthop- 
terygian fishes has also been identified in gobioids. These two systems are known 
to occur together in spiny-rayed fishes only in percopsiforms. These important 
similarities of the nerves, together with those summarized by McAllister (1968), 
add strongly to his suggestion that these fishes should stand in systematic re- 
lationship to the percopsiform fishes. It is proposed they be given ordinal status, 
the Gobiiformes, with characters as given in McAllister (1968), Regan (1911), 
and with the nerve and sensory canal features given in the present paper. 

ADDENDUM 

The excellent paper by Rosen and Patterson (1969) on the structure and 
relationships of paracanthopterygian fishes was received too late to be of use in 
the present study. The systematic results in both papers are mostly in agree- 
ment. Four differences are noted. One is the addition of gobioid fishes to the 
Paracanthopterygii. Rosen and Patterson state that they could not find any 
other group that should be included. Gobioids lack the caudal fin structure of 
percopsiforms but some {Dormitator and Gunnellichthys) have what may be a 
levator maxillae superioris muscle (W. Eschmeyer, unpublished information). 
So do the Sciaenidae, but here it is evidently convergent. The Gobiesocidae have 
neither of these features but are included by Rosen and Patterson on other evi- 
dence. The gobies qualify by the nerves and numerous other similarities to 
percopsids which together add weight to an argument for percopsid relation- 
ships. Another difference concerns the placement of the Amblyopsidae. Rosen 
and Patterson put them in the same suborder with the Aphredoderidae. The 
occurrence of the same unique pattern of RLA in both the Amblyopsidae and 
Percopsidae, and of a very different pattern in Aphredoderus resembling those 
of other paracanthopterygians and of acanthopterygians, would seem to outweigh 
the similarities Rosen and Patterson used for their placement of amblyopsids. 
The question needs further study. Maybe no suborders are needed here. Another 
difference concerns the origin of the Paracanthopterygii, needless to say a diffi- 
cult question on which to have good information. In the present paper an os- 
merid ancestry is hypothesized for the percopsiforms. Rosen and Patterson 
thought (as of going to press in 1969) that important parallel developments of 
myctophoids to paracanthopterygians and acanthopterygians do not indicate 
that paracanthopterygians originated from myctophoid fishes. They give these 
three groups superordinal rank representing three "parallel radiations into a neo- 
teleostean grade" (Rosen and Patterson, 1969). They state further that "poly- 



260 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

mixioids can tentatively be viewed as the closest relatives of the paracanthoptery- 
gians." Figure 20 of the present paper agrees in general with their superordinal 
relationships but it shows a percopsiform ancestry for the acanthopterygii and 
an osmerid ancestry for the percopsid fishes. Nerve evidence and an intuitive 
finger point in an osmerid direction. In the search for percopsiform ancestors 
the guidelines of levator maxillae superioris muscle and percopsiform caudal fin 
structure break down. New guidelines have to be looked for and followed. These 
at present are nerves. An exciting problem emerges as to what living fishes may 
be closest to the ancestors of the percopsids. 

A final point of difference to be mentioned is that Rosen and Patterson 
suggest that primitive percopsiforms had sensory head canals enclosed in bone 
and spinous suborbitals and preopercle and that these features were reduced 
and lost in paracanthopterygian but not acanthopterygian derivatives. Their 
evidence for these assumptions is slim. An osmerid ancestry makes it unneces- 
sary to build closed canals. None of the percopsiform fishes living or fossil, have 
enclosed canals except for a very small part at the posterior end of the supra- 
orbital canal in Percopsis. Osmerids have open canals in all bones; smooth sub- 
orbitals, lacrymal, and preopercle; an antorbital; an adipose fin; the ramus ca- 
nalis lateralis system less specialized; and ventral segmental lateralis nerves less 
specialized than in Percopsis. Closed and spinous canals would have come after- 
ward in percopsiform lines leading to beryciform fishes. 

Evidence in support of the Paracanthopterygii is now substantial and comes 
from bones, muscle, nerves, and fossils. It provides new understanding of fish 
evolution. We should thank Dr. Rosen and his coworkers for their achievements. 

LITERATURE CITED 

Agassiz, Louis I., Elliot Cabot, and others 

1850. Lake Superior, Boston, 428 pp., 17 pis. 
Freihofer, Warren C. 

1960. Neurological evidence for the relationships of some percomorph fishes. 196 pp. 

Doctoral dissertation, Stanford University, Stanford, California. 
1963. Patterns of the ramus lateralis accessorius and their systematic significance in 
teleostean fishes. Stanford Ichthyolop:ical Bulletin, vol. 8, no. 2, pp. 79-189, 29 
figs., 2 tables. 
1966. The Sihler technique of staining nerves for systematic study especially of fishes. 
Copeia, 1966, no. 3, pp. 470-475, 2 figs. 
Frost, G. .'Illen 

1925. A comparative study of the otoliths of the neopterygian fishes. .Annals and Maga- 

zine of Natural History, ser 9, vol. 15, pp. 152-163, pis. 11-13. 

1926. A comparative study of the otoHths of the neopterygian fishes (continued). ■ — 

Orders Haplomi, Heteromi, Iniomi, Lyomeri, Hypostomides, Salmopercae, 
Synentognathi, Microcyprini, Solenichthys. Annals and Magazine of Natural 
History, ser. 9, vol. 18, pp. 465-482, pis. 20-21. 
192 7. A comparative study of the otoliths of the neopterygian fishes (continued). 
Annals and Magazine of Natural History ser. 9, vol. 20, pp. 298-305, pi. 5. 



Vol. XXXVIII] FREIHOFER: XERVE PATTERNS 261 

GOSLINE, WiLLUM A. 

1968. The suborders of perciform fishes. Proceedings of the United States National 
Museum, vol. 124, no. 3647, 78 pp., 12 figs., 3 tables. 
Greenwood, P. Humphrey, D. E. Rosen, S. H. WEixzitAN, and G. S. Myers 

1966. Phyletic studies of teleostean fishes, with a provisional classification of living 
forms. Bulletin of the American Museum of Natural Histor>^, vol. 131, art. 4, 
pp. 339-456, pis. 21-23, 32 charts, 9 text figs. 
H.ANDRicK, Kurt 

1901. Zur kenntnis des Nervensystems und der Leuchtorgane des Argyropeleciis hem- 
igymnus. Zoologica, Heft 32, 68 pp., 6 pis., Stuttgart. 
McAllister, D. E. 

1968. Evolution of branchiostegals and classification of teleostome fishes. Bulletin 221, 

Biological series no. 77, National Museum of Canada, i-xiv + 239 pp., 2 
tables, 21 plates, 3 figs. 
Manigk, Wolfgang 

1934. Der tngemino-facialiskomplex und die Innervation der Kopfseitenorgane der 
Elritze (Phoxinus laevis). Zeitschrift fiir Morphologic und Okologie der 
Tiere, vol. 28, pp. 64-106, 16 figs. 
R.\v, D. L. 

1950. The peripheral nervous system of Lampanyctus leucopsarus. Journal of Mor- 
phology, vol. 87, no. 1, pp. 61-178, 40 figs. 
Regan, C. Tate 

1911. The osteolog>- and classification of the gobioid fishes. Annals and Magazine of 

Natural History, ser. 8, vol. 8, pp. 729-733, 2 figs. 
1913. The classification of percoid fishes, .\nnals and Magazine of Natural History, 
ser. 8, vol. 12, pp. 111-145. 
Rosen, Donn E. 

1962. Comments on the relationships of the North .American cave fishes of the family 
Amblyopsidae. American Museum Novitates, no. 2109, 35 pp., 24 figs. 
Rosen, Donn E., and Colin Patterson 

1969. The structure and relationships of the paracanthopterygian fishes. Bulletin of the 

American Museum of Natural History, vol. 141, article 3, pp. 357-474, figures 
1-74, plates 52-78, tables 1-8. 
Weitzman, Stanley H. 

1967. The origin of the stomiatoid fishes with comments on the classification of sal- 
moniform fishes. Copeia, no. 3, pp. 507-540, 18 figs. 

LIST OF ABBREVIATIONS 

BBP. — branch from brachial plexus. 

BDL-L. — Baudelofs ligament. 

BLLOV. — branch to lateral-line nerve and opercular nerve of vagus. 

B-OCC.-SP-CO. — branches to occipito-spinal complex. 

BVSR. — branch to ventral spinal ramus. 

CL. — cleithrum. 

CM. — cartilaginous membrane. 

DRSN. — dorsal ramus of spinal nerve. 

EFRN. — external fin-ray nerve. 

F-LAT-OR. — free, naked lateralis organ. 

FDR-OCC-SP. — first dorsal ramus of occipito-spinal complex. 



262 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

FR. — fin ray. 

GG. — geniculate ganglion. 

HYO. — hyomandibular. 

IDR-OCC-SP-CO. — first dorsal ramus of occipito-spinal complex. 

LLN. — lateral-line nerve. 

NINTH. — glossopharyngeal nerve. 

OBRN. — opercular branch from vagus nerve. 

OCC-SP-N&G — occipito-spinal nerves and ganglia. 

OP. — opisthotic. 

PAR. — parietal. 

PEC-PEL. ACC. — ventral pectoral-pelvic accessory lateral-line nerve. 

PHBRV. — pharyngo-branchial nerves of vagus. 

PP-ACC-LAT. — ventral, segmental accessory lateral-line nerves. 

PRF. — prootic facial foramen shown as dashed black line medial to hyomandibular. 

PRO.— prootic. 

PTR. — pterotic bone. 

PTR-SP. — pterotic spine. 

PTT. — posttemporal. 

RCIA. — supraorbital branch of ramus canalis lateralis system. 

RCIB. — temporal branch of ramus canalis lateralis system. 

RC2. — anterior infraorbital branch plus dorsoanterior dentary and rictus branches of ramus 

canalis lateralis system. 
RC3, RC3A, RC3B. — posteroventral dentary branches plus rictus branches of ramus canaUs 

lateralis system. 
RC3C. — branch of ramus canalis lateralis system detaching from ramus opcrcularis super- 

ficialis facialis and extending to anteroventral area of preopercular canal membrane. 
RC4, RC4A. — preopercular branch of ramus canalis lateralis system. 
RC4B. — medial preopercular-mandibular ridge prolongation of RC4A. 
RCS. — supratemporal branch of ramus canalis lateralis system. 
RCL. — ramus canalis lateralis nerve. 
R-HYO. — ramus hyoideus. 

RLA. — ramus lateralis accessorius or recurrent facial nerve. 
RLA-A. — branch of RLA to anal fin. 
RLA-D. — branch of RLA supplying dorsal fin. 
RLA-DFR.— branch of RLA to dorsal fin ray. 
RLA-OCC-SP-CO. — branches of RLA to occipito-spinal complex. 
RLA-P. — branch of RLA to pectoral fin. 

RLA-PDA. — a main branch of RLA extending from parietal to dorsal fin. 
RLA-PEL.— branch of RLA to pelvic fin. 
RLA-PP. — a main division of RLA extending from parietal bone to pectoral and pelvic 

fins. 
RLA + SYM.- — common trunk of RLA -f- sympathetic. 
RLA + SYM-DASH. — ramus lateralis accessorius plus sympathetic common trunk shown 

as dashed black line medial to hyomandibular. 
RLA-THYO — a main division of RLA extending from prootic foramen to pectoral and pelvic 

fins. 
R-MAN-EX. — ramus mandibularis externus. 
R-MAN-IN. — ramus mandibularis internus. 
SPH. — splenotic. 



Vol. XXXVIII] FREIHOFER: NERVE PATTERNS 263 

SPN. — spinal nerve. 

SUPCL. — supraclei thrum. 

SYM. — sympathetic trunk. 

SYM-G. — sympathetic ganghon. 

TR-COM-SYM. — transverse commissure of sympathetic. 

TR-HYO. — truncus hyomandibularis. 

VR-OCC-SP-CO. — ventral ramus of occipito-spinal complex 

VSN. — branch of ventral spinal nerve. 

VSR. — ventral spinal ramus. 



PROCEEDINGS 

OF THE 

CALIFORNIA ACADEMY OF SC:iENCES 

FOURTH SERIES 



Festschrift for George Sprague Myers 



Vol. XXXVIII, No. i:^ |)j>. 2fi5-272; 2 figs.; 3 tables. December .^1, 1970 



A NEW GYMNOTOID FISH FROM THE 
RIO TOCANTINS, BRAZIL 

By 

Maarten Korringa 

nivisott of Sysletnalir Itiology, Stanford Vniversily 

IN'IRODIK 'IION 

Amonf^st the f^ymncjtoids in the fish collection of the C.'uilfornia Academy (jf 
Sciences in vSan Francisco, I found three unusual specimens which were collected 
in 1924 at I'orto Nacional, Rio Tocantins, Brazil, by Dr. Carl Ternetz. Accord- 
ing^ to the keys in Kllis (1913) and Schultz (1949), these specimens fall into the 
genus Strrnopvi^us, but the unusual appearance of the new fish suggested that it 
might be different. Accordingly, the three specimens were radiographed, and, 
after morphomelric data were taken, one was cleared and stained for osteological 
study. Although these fish superficially resemble Stcrnupygus, their affinities do 
not appear to lie with this genus, i believe that the .species described below re- 
quires generic recognition. Osteological nomenclature follows Weitzman (1962). 

ACKNOWLEDGMENTS 

1 am grateful to Dr. W. I. l-ollett and Mrs. Robert Dempster of the Califor- 
nia Academy of Sciences in San l^ranci.sco for their kind assistance in obtaining 
study materials from the fish collection there. F am also indebterl to Professor 
George S. Myers, Mr. Leonard Com[)agno, and Dr. Warren C. Freihofer for their 
encouragement and criticism of the work at various stages. 

|26,S| 



266 



CALIFORNIA ACADEMY OF SCIENCES 



[Proc. 4th Ser. 



Vol. XXXVIII] KORRINGA: GYMNOTOID FISH FROM BRAZIL 



267 







zz V- 







5mm 



Figure 2. Archolaemus blax, holotype. View of jaws, indicating shape of tooth patches 
and showing mandibular teeth outside mouth. 2a. Upper jaw ; 2b. lower jaw. 



Archolaemus Korringa, new genus 

Type species. Archolaemus blax Korringa, new species. 

Elongate compressed gymnotoid fishes, lacking caudal fin and dorsal thong. 
Frontal and parietal fontanels large. Tail extending beyond end of anal fin. Body 
completely covered with scales, with several rows of large scales near and below 
lateral line; back and lower parts of sides covered with very small scales. Head 
naked. Orbital margin free. Mesopterygoid with approximately 10 small villi- 
form teeth. Premaxillaries and dentaries each carrying numerous villiform teeth 
in a single patch on each bone (see fig. 2) ; a number of dentary teeth are out- 
side the mouth and project forward in larger specimens. Premaxillaries small, 
about one and one-half times as long as wide, about one-fourth as long as the 
maxillaries. Maxillaries more or less straight, virtually parallel to the body 
length of the fish. Branchiostegal rays 5, 4 on ceratohyal, 1 on epihyal. Posterior 
process of the lower pharyngeal bones with a patch of approximately 12 teeth. 
The upper pharyngeals each with a patch of about 8 teeth. In both upper and 
lower pharyngeals these teeth are small and villiform. Mesocoracoid absent. 

Certain lateral line canals of the head are much larger in diameter than others, 
and larger even than the canal of the body. Especially prominent are the nasal- 
supraorbital canal (anterior to the eyes), the infraorbital canal (excepting the 
posterior 2 bones, which are of small diameter), and the preopercular-mandibular 
canal. In conjunction with this increase in diameter, there is a progressive re- 
duction of the superficial walls of the canals, so that they resemble troughs roofed 
over by arches. 



268 



CALIFORNIA ACADEMY OF SCIENCES 



IProc. 4th Ser. 



Table 1. Measurements in millimeters. Figures in parentheses are measurements ex- 
pressed as percentage of the distance from tip of snout to end of anal fin base. 





Holotype 
CAS 24743 


Paratype 
CAS 24744 


Paratype 
CAS 24745 


Total length 


435 


(127) 


313 


(124) 


235 


(137) 


Snout tip to end of anal fin 


344 


(100) 


2Si 


(100) 


172 


(100) 


Snout tip to origin of anal fin 


46 


(13.4) 


45 


(17.8) 


26.5 


(15.4) 


Longest anal fin ray 


16.5 


(4.8) 


14.5 


(5.7) 


8.5 


(4.9) 


Snout tip to vent 


2i 


(6.7) 


18 


(7.1) 


17 


(9.9) 


Snout tip to occiput 


33.5 


(9.7) 


30 


(11.9) 


20 


(11.6) 


Snout tip to pectoral fin origin 


47 


(13.7) 


45 


(17.8) 


25.5 


(14.8) 


Length of pectoral fin base 


8 


{2.i) 


6.5 


(2.6) 


3.5 


(2.0) 


Snout tip to tip of pectoral fin 


7S.S 


(21.9) 


71 


(28.1) 


42 


(24.4) 


Snout tip to anterior margin of eye 


19 


(5.5) 


18 


(7.1) 


11 


(6.4) 


Snout tip to rictus^ 


6 


(1.7) 


— 


— 


3.5 


(2.0) 


Snout tip to posterior edge of opercle 


43.5 


(12.7) 


40 


(15.8) 


23 


(13.4) 


Snout tip to anterior nostriP 


4.3 


(1.3) 


— 


— 


2.7 


(1.6) 


Posterior nostril to eye 


13.5 


(3.9) 


11.5 


(4.5) 


6 


(3.5) 


Distance between orbital margins 


4.5 


(1.3) 


3 


(1.2) 


2.6 


(1.5) 


Depth at eye 


22.5 


(6.5) 


19 


(7.5) 


12.5 


(7.3) 


Depth at occiput 


29 


(8.5) 


25 


(9.9) 


15.5 


(9.0) 


Maximum depth of body- 


40.5 


(11.8) 


40 


(15.8) 


19.5 


(11.3) 


Width of mouth 


6 


(1.7) 


5.5 


(2.2) 


2.7 


(1.6) 


Width of head at eye 


13.5 


(3.9) 


11 


(4.3) 


6 


(3.5) 


Width of head at occiput 


15.5 


(4.5) 


14 


(5.5) 


9 


(5.2) 


Anal fin rays 




218 




02 


202 


Pectoral fin rays, right side 




19 




20 




20 


Pectoral fin rays, left side 




19 




20 




19 


Lateral hne scales" 




137 


143 


146 


Abdominal vertebrae 




IS 




14 




14 


Caudal vertebrae 




59 




51 




75 



1 Snout of this specimen too severely damaged to take accurate measurements. 

- Maximum depth of body is approximately at tip of pectoral fins. 

^ Counted between dorsal margin of gill opening and end of anal fin base. 



Abdominal vertebrae 14 or 15; caudal vertebrae 51 to 75 in the three speci- 
mens; actual range doubtless greater. Two pyloric caeca present. Vent and 
genital papilla lie directly below^ the eye. Snout moderately long, conical; dis- 
tance from snout tip to eye approximately equal to distance from eye to posterior 
margin of opercle. 

Archolaemus blax Korringa, new species. 

(Figures 1, 2.) 

Study material. Three specimens. Holotype: CAS 24743, male; 435 mm. in 
total length; Porto Nacional, Rio Tocantins, Estado de Goias, Brazil; collected 
by Carl Ternetz, February 8, 1924. Paratypes: CAS 24744, female full of eggs, 



Vol. XXXVIin KORRIXGA: GYMNOTOID FISH FROM BRAZIL 269 

313 mm.; and CAS 24745, 235 mm. (cleared and stained with alizarin, in glyc- 
erine), both collected with the holotype. 

Description. See table 1 for counts and measurements. Dorsal profile of 
head straight: ventral profile slightly convex to slightly concave. Tail somewhat 
flattened and ribbonlike (this may be an artifact, as all the tails possess, instead 
of vertebrae, a slender flexible rod, indicating that regeneration has taken place). 
One paratype (CAS 24744) is a ripe female; the eggs are slightly ovoid, about 
1.5 mm. in length. The first gill arch of the right side possesses 7 to 10 nubbins 
representing rakers (9 in holotype, 10 and 7 in paratypes). In Sternopygus each 
nubbin has imbedded in it a number of very small spines or teeth. I have not 
been able to determine if this is true for Archolaemus, as the spines or teeth are 
not present in the cleared and stained specimens. 

Color in alcohol an even tan; area around orbital margin and tip of snout 
pale. There is no evidence of the dark brown banding found in many gymnotoids. 

Origin of name. Greek, archos, anus; laimos, throat, from the location of 
the vent under the eye; Latin, Max, doltish, in reference to the fish's general 
appearance. 

DISCUSSION 

Regan (1911) places all short-snouted gymnotoids with frontal and parietal 
fontanels, but lacking caudal fins and dorsal thongs, in the subfamily Sternopy- 
ginae of his family Sternarchidae (properly Apteronotidae). I assigned Archo- 
laemus to this subfamily on the basis of 1 ) absence of mesocoracoid, 2 ) presence 
of large fontanels, 3) absence of caudal fin and dorsal thongs, 4) presence of 
mesopterygoid teeth. Sternopygus and Archolaemus are the only members of the 
subfamily with a free orbital margin. However, Archolaemus is distinct from 
Sternopygus in several ways. Sternopygus has a pectoral girdle with a short 
suture between the coracoid and the scapula, the scapular foramen being open 
anteriorly. In Archolaemus, the suture between the two bones is considerably 
longer, and a distinct scapular foramen is present. The former type, as Regan 
points out, is found in Steatogenys, whereas the latter is characteristic of Eigen- 
mannia. The pectoral girdle of Hypopomus resembles that of Eigenmannia and 
Archolaemus in possessing a long suture, but lacks a conspicuous scapular fora- 
men. Archolaemus is further distinguished in having the anus and genital pa- 
pilla directly below the eye; in all specimens of Sternopygus examined, these 
were well posterior to the eye. The eye of Archolaemus is considerably larger 
(about 4.5 to 5.5 times in the distance between the eye and the posterior margin 
of the opercle) than that of Sternopygus (6.5 to 11). Other distinguishing 
features are: premaxillaries small, about one-fourth the length of the maxil- 
laries; as opposed to slightly over one-half in S. macrurus; few anal rays, 
generally less than 220, whereas Sternopygus has 234 to 320 (counts between 
260 and 280 are typical). The maxillaries are fairly long and almost horizontal. 



270 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

whereas Sternopygus has obUque maxillaries. The body cavity of Archolaemus 
is short (6.3 to 8.3 times in length to end of anal fin base), and few (14 to 15) 
abdominal centra are present; Sternopygus has a longer body cavity (about 3.5 
to 5 times in length to end of anal fin base) and more abdominal centra (gener- 
ally 20 to 25). Archolaemus has a much longer snout than Sternopygus, the 
distance from snout tip to eye being about 1.2 times the distance from eye to 
occiput, as opposed to 0.7 for the latter genus. 

Archolaemus blax is distinguished from all species of Eigenmannia by the 
presence of a free orbital margin. Except for E. virescens, no species of Eigen- 
mannia that I have examined has the vent as far forward as the eye. Archo- 
laemus has a significantly longer snout than does Eigenmannia; snout tip to 
center of eye is contained 13 to 16 times in the length to the end of anal fin base 
in the former, 20 to 28 times in members of the latter genus. Archolaemus has a 
long maxillary disposed almost parallel to the body length, whereas the maxil- 
laries of Eigenmannia are much shorter and vary from 45° to perpendicular to 
the body length. Archolaemus appears to have a proportionately longer gape 
than Eigenmannia. No specimen of Eigenmannia that I have examined has 
teeth outside the mouth. 

Rhabdolichops (Eigenmann and Allen, 1942) is distinguished from Archo- 
laemus principally by its squamation, gill rakers, and general body shape. The 
back of Rhabdolichops is naked to a point about two-thirds of the distance to 
the end of the anal; anteriorly, the entire region above the lateral line is naked; 
posteriorly, only the top of the back. The body of Archolaemus is entirely 
covered with scales. Rhabdolichops has long, well developed gill rakers, a short 
snout, and a concave upper head profile. I do not yet have osteological material 
of this genus and therefore am not certain of its affinities. 

The edentulous Steropyginae {Hypopomus, Steatogenys, and Parupygus) 
appear to be highly distinct from the foregoing genera. Most notably, Archo- 
laemus, Sternopygus, Eigenmannia, and Rhabdolichops have similar lateral line 
canals while in Steatogenys, Hypopomus, and Parupygus (Hoedeman, 1962) all 
the canals are tubelike and of small diameter. The circumorbital tubes lack 
platelike stays. Many of the canals seem to be isolated from the bones with 
which they are normally associated in other groups of fishes; e.g., there is no 
apparent bony connection between the dentary and the chain of tubules lying 
ventral to it. Furthermore, these 3 genera lack teeth on the premaxillaries, den- 
taries, and mesopterygoids and not one has a long snout. Archolaemus further 
differs from Steatogenys in the absence of "mental filaments" and lacks the 
banded color pattern found in this genus and in some species of Hypopomus. 

COMPARATIVE MATERIAL EXAMINED 

Sternopygus macrurus (Bloch and Schneider): SU 21997, 2 specimens. Bo- 
tanic Garden, British Guiana. Eigenmannia virescens (Valenciennes) : SU 



Vol. XXXVIII] KORRIXGA: GYMNOTOID FISH FROM BRAZIL 271 

54508, 2 specimens, Lagoa Grande, Brazil. Eigenmannia macrops (Boulenger): 
SU 54473, 2 specimens, Sao Gabriel Rapids, Rio Negro, Brazil. Eigenmannia 
conirostris Eigenmann and Allen: SU 54461, 1 specimen, Lagoa Grande, Lower 
Amazon, Brazil, in alcohol. Rhabdolichops longicaudatus Eigenmann and Allen: 
SU 54377, 1 specimen, Santerem, Amazon, Brazil; SU 64076, 1 specimen, Cu- 
cuhy, Rio Negro, Brazil; both in alcohol. Hypopomus brevirostris (Steindach- 
ner): SU 24769, 1 specimen, Lago Gatiin, Three Rivers Plantation, Panama. 
Steatogenys elegans (Steindachner) : SU 22445, 2 specimens, Belem do Para, 
Brazil. Pariipygiis savanensis Hoedeman: Zoological Museum of Amsterdam 
106,074, 1 specimen, Botopasi, Surinam River, Surinam. Except as noted, all 
are alizarin preparations in glycerine. 

SUMMARY 

Archolaemus blax Korringa, a new genus and species of gymnotoid fish is 
described. It is one of the toothed Sternopyginae, a group consisting of Sterno- 
pygus, Eigenmannia., and Rhabdolichops. These are united by the possession of 
hyper trophied lateral Hne canal bones in the head. Archolaemus shares many 
characters with Sternopygus and Eigenmannia, though its affinites lie more with 
the latter. 

LITERATURE CITED 

Eigenmann, C. H., and W. R. Allen 

1942. Fishes of Western South America. Lexington, Kentucky, xv + 494 pp., 22 pis., 
1 map. 
Ellis, M. M. 

1913. The gymnotoid eels of tropical America. Memoirs of the Carnegie Museum, vol. 
6, no. 3, pp. 109-204, pi. 19-23. 
Hoedeman, J. J. 

1962. Notes on the ichthyology of Surinam and other Guianas, 9. New records of gym- 
notid fishes. Bulletin of Aquatic Biology, vol. 3, no. 26, pp. 53-60. 
Regan, C. T. 

1911. The classification of the teleostean fishes of the order Ostariophysi, 1. Cypri- 
noidea, Annals and Magazine of Natural History, ser. 8, vol. 8, no. 43, pp. 
13-31, pi. 2. 

SCHULTZ, L. p. 

1949. A further contribution to the ichthyology of Venezuela. Proceedings of the 
United States National Museum, vol. 99, pp. 1-211, pi. 1-3. 
Weitzman, S. H. 

1962. The osteology of Brycon meeki, a generalized characid fish, with an osteological 
definition of the family. Stanford Ichthyological Bulletin, vol. 8, no. 1, pp. 1-77. 



PROCEEDINGS 

OF THE 

CALIFORNIA ACADEMY OF SCIENCES 

FOURTH SERIES 



Festschrift for George Sprague Myers 



Vol. XXXVIII, No. 14, pp. 273-288; 3 figs. December 31, 1970 



ON THE TRAIL OF THE GOLDEN FROG: 

WITH WARSZEWICZ AND GABB 

IN CENTRAL AMERICA 

By 

Jay M. Savage 

Defjartntent of Biological Sciences, Allan Hancock Foundation, 
University of Southern California 

Those who have viewed at first hand the steep, dark-green, forest-covered 
slopes of the Cordillera de Talamanca-Chiriqui of Costa Rica and Panama, with 
their ever changing aspect of sun and cloud, moon and mist, bright blue sky and 
bright green mantle, driving rain and boiling fog, come away with a feeling of 
overpowering awe and mystery at the variety of nature and the magic of the 
human soul. It is not surprising that the primitive peoples in this region also 
regarded the mountains and their forests with mystical reverence, so near and yet 
towering abruptly upwards to 4,000 meters from their lowland valley habitations. 

Among the Bribri, Cabecar, Boruca, Changina, and Chiriqui, when the chicha 
has been drunk, the night grows late and dark, and the fires die down to burning 
embers, the wisest old man of the tribe tells his engrossed listeners of a beautiful 
miraculous golden frog that dwells in the forests of these mystical mountains. 
According to the legends, this frog is ever so shy and retiring and can only be 
found after arduous trials and patient search in the dark woods on fog shrouded 
slopes and frigid peaks. However, the reward for the finder of this marvelous 
creature is sublime. Anyone who spies the glittering brilliance of the frog is at 
first astounded by its beauty and overwhelmed with the excitement and joy of 
discovery; almost simultaneously he may experience great fear. The story contin- 

[273] 



274 



CALIFORNIA ACADEMY OF SCIENCES 



IProc. 4th Ser. 




^ 



^l<y^ 




^^ 




Figure 1. Upper, Josef Warszewicz, orisinal by Artura Grottgera, now the property of the 
Department of Plant Geography, Jagellonian University, Krakow; lower, William M. Gabb, 
original, the property of the Academy of Natural Sciences of Philadelphia. 



Vol. XXXVIII] SAVAGE: WARSZEWICZ AND GABB 275 

ues that any man who finds the legendary frog finds happiness, and as long as 
he holds the frog happiness will follow him everywhere. The story tellers record 
many men who have scaled the highest peaks and searched the darkest forests 
for even a glimpse of the golden frog, but only a few ever see it. Fewer still cap- 
ture the cherished creature and hold him for a few moments, and a very few are 
able to carry him with them for a longer period of time. One story tells of the 
man who found the frog, captured it, but then let it go because he did not recog- 
nize happiness when he had it; another released the frog because he found happi- 
ness too painful. 

Like the Indians of Talamanca and Chiriqui, each human being is also on a 
mission searching for the golden frog. Field biologists in particular seem always 
to be searching for mystical truth and beauty in nature, and frequently at some 
unperceived level, for that happiness promised by the Indian seers. The present 
paper is appropriately about two 19th Century scientists who joined this search 
in the very regions where the golden frog abounds, and we may assume that for 
a time, at least, they captured that joy guaranteed to beholders of the frog. 

INTRODUCTION 

Both Josef Warszewicz and William M. Gabb (fig. 1) were pioneer collectors 
of herpetological materials from lower Central America. Since they were among 
the first to sample the region, many of the animals they collected became types 
of previously undescribed species, most of which remain recognized as valid today. 
Neither of these men was a zoologist, and both collected in regions not visited 
again by herpetological collectors until the present century. Confusion and doubt 
as to the origin of their collections have clouded the issue of the validity of certain 
names and the synonymy of others subsequently described. In the present paper 
the routes followed by the two pioneers and the sources of their materials are 
delineated for the first time. 

ACROSS THE GREAT DIVIDE: 
WARSZEWICZ IN WESTERN PANAMA 

Josef Warszewicz was born in Litwie (Wilno) , Poland in 1812. He apparently 
studied some botany at the University of Krakow. He took part in the Polish 
Revolution against Russia of 1830-31 and rose to the rank of officer. After the 
defeat of the Polish insurgents he left Poland. From 1840-1844 he worked as a 
gardener in the Botanical Gardens in Berlin. There he came to the attention of a 
Belgian, Van Houtte de Gandawy, who owned a large garden in Santo Tomas (St. 
Thomas), now Matias de Galvez, Guatemala. Warszewicz was sent to inventory 
the garden and to collect materials for Belgian gardens. He sailed from Europe, 
December 5, 1844, and was active in Guatemala by March 1, 1845. He added 
many local species to his employer's gardens and in 1846 began work for himself, 
and forwarded living and dry plants, especially orchids, to Europe. In 1848 



276 



CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 




Figure 2. Map of western Panama, showing Warszewicz' route across the Continental 
divide and principal localities discussed in text. 



Vol. XXXVIII] SAVAGE: WARSZEWICZ AND GABB 277 

Warszewicz undertook a major trip through Central America. He traveled by 
land from Guatemala to San Jose, Costa Rica, where he was situated by February, 
1848. On ]March 1, he climbed \'olcan Irazu. Later he arrived in western 
Veragua (Chiriqui), Panama, where he climbed Volcan Chiriqui and crossed 
over to the Caribbean coast. Most of the amphibians and reptiles collected by 
Warszewicz were taken in western Panama. In 1851 he was again in the Chiriqui 
region and later that year he proceeded to South America, being in Guayaquil, 
Ecuador, at the end of the year. Warszewicz spent 1852 in South America, pri- 
marily in Peru and Bolivia. He is known to have visited Lima, Peru, and was at 
La Paz, Bolivia, on June 15, 1852. At the end of the year, December 28, he was 
at Huancabamba on the headwaters of the Rio IMarahon, upper Amazon drainage, 
Departamento Piura, Peru. He returned to Germany in October 1853 and became 
Inspector of the Botanic Gardens in Krakow. He died there December 29, 1866. 
(Regal, 1867; Rouppert, 1927) . A bust of Warszewicz was erected in the Univer- 
sity Botanical Garden in Krakow about 1880, where it still stands. 

Herpetological materials collected by Warszewicz were deposited at Berlin, 
Vienna, and Krakow. The last city had been made the capital of a small free 
state after the Napoleonic wars in 1815. In November 1846, it was annexed to 
the Austrian-Hungarian Empire, following a revolt in Poland. Through exchange, 
some specimens came to the museum at Munich and to the British Museum. The 
Central American specimens are all from Panama and were taken in 1848 and 
1851. The former collection seems to have gone to Krakow and Vienna, the last 
to Berlin. The long residence of Warszewicz in Berlin prior to his American 
travels explains deposition of specimens there, probably as the result of long-time 
contacts. Apparently he loaned and gave some material to the \"ienna ISIuseum 
on his establishment in Krakow in 1843, since the latter city was then part of the 
Austrian-Hungarian state. Fortunately Warszewicz' route through the Chiriqui 
massif may now be traced with some accuracy (fig. 2). Information provided by 
Wagner (1863) and his map clearly define the route from David on the Pacific 
slope across the divide to the Laguna de Chiriqui. Wagner records that Warsze- 
wicz penetrated the interior of Chiriqui and traversed the great Cordillera to the 
Atlantic shore. Kegel (1867) noted that Warszewicz climbed the 16,000 foot 
Volcan Chiriqui in 1848. Wagner (1863), using the same guides and carriers 
employed by Warszewicz, followed the same trails to the Chiriqui highlands. 
This route runs from David through Dolega and then up to Boquete (1158m.), 
from where there were two trails leading to the Caribbean shore. One trail 
skirted the east slope of Volcan Chiriqui and continued to Ranchos de Robalo, the 
other passed around the east slope of Cerro Horqueta and continued to the mouth 
of Cabbage Creek (Rio Guarmo) near present day Chiriqui Grande. Warszewicz 
certainly followed the Boquete-Robalo trail, since a branch from it leads to the 
top of Volcan Chiriqui (3478m.). Nevertheless, he may have returned via the 
other route. Emmett R. Dunn and Chester B. Duryea seem to have followed the 



278 CALIFORNIA ACADEMY OF SCIENCES IProc. 4th Ser. 

latter trail from Chiriqui Grande to Boquete, in 1923 when they became the first 
herpetologists to re-collect several of Warszewicz' species. 

HERPETOLOGICAL SPECIMENS COLLECTED 
BY WARSZEWICZ 

Most of the amphibians collected by Warszewicz were described by Oskar 
Schmidt (1857) and more extensively described and illustrated by him in 1858. 
The following are involved, with the Krakow Museum in the Department of 
Systematic Zoology, Jagellonian University (KM) numbers listed. These speci- 
mens were presented to the collection in 1870 and were examined by E. R. Dunn 
in 1928. Some of them are still extant. Location of other types and other Wars- 
zewicz material noted in preparation of this report is also indicated, but probably 
is not complete. Abbreviations for other collections are: Zoologisches Museum, 
Berhn (B); British Museum (Natural History), (BM); Zoologischen Museum, 
Hamburg (H); Zoologischen Staatssammlung in Miinchen (M); Naturhistor- 
isches Museum Wien (W). An asterisk (*) indicates a new species. 

Specimens Collected by Warszewicz 

*Leiuperus sagittifer. New Granada (Colombia). 

*Ixaliis warschewitschii. KM 1006/1338; near Volcan Chiriqui, between 6000 and 7000 feet 

(4500-5250 feet = 1370-1600 m.). 
*Hyla piignax. KM 1009/1339; Rio Chiriqui near Bocas del Toro. 
*Hylasplendens. KM 1008/1340 9 :Rio Chiriqui near Bocas del Toro. 
*Hyla molitor. KM 1010/1341, 2 c? (^ ; W 16494, female designated as lectotype by Savage 

and Heyer (1969) :Rio Chiriqui near Bocas del Toro. 
*Hyla molitor marmorata. KM 1010/1342 9 :Rio Chiriqui near Bocas del Toro. 
*Hylodes fitzingeri. KM 1012/1343; Mountains of New Granada (Panama), 4000 feet 

(3000 feet = 915 m.) ; now lost. 
*Dendrobates speciosns. KM 1017/1345 nine specimens; W one specimen: trail between 

Bocas del Toro and Volcan Chiriqui, 5000-7000 feet (3777-5250 feet = 1150-1600 m.) ; 

now lost. 
* Dendrobaies pumilio. KM 1018/ 1346: trail between Bocas del Toro and Volcan Chiriqui, 

5000-7000 feet (3777-5250 feet = 1150-1600 m.) ; now lost. 
* Dendrobaies lugubris. KM 1016/1347: trail between Bocas del Toro and Volcan Chiriqui, 

5000-7000 feet (3777-5250 feet = 1150-1600 m.) ; now lost. 
Bufo margaritifer. Between Bolivia and Peru, 3000 feet (2250 feet = 685 m.). 
*Bufo pleuropterus. KM 1030/1348: between Bohvia and Peru, 3000 feet (2250 feet = 685 m.). 
*Bufo veraguensis. KM 1032/1350; New Granada, Provincia de Veragua. 
*Bufo simus. BM 95-9-14.6; H 1527; KM 1029/1351, 5 specimens (now lost); M 543/20; 

W 16521: Rio Chiriqui near Bocas del Toro. 
* Hylaemorphtis dumerilii. KM 1014/1345: New Granada, Provincia de Chiriqui, 8000 feet 

(6000 feet = 1830 m.). 
*Hylaemorphus bibronii. KM 1015/1355; New Granada near Panama, 2000-3000 feet (1500- 

2500 feet = 460-760 m.). 
*Phirix pachydermus. KM 1013/1356; Western New Granada near Buenaventura, 5000 feet 

(3777 feet z=1150m.) ; now lost. 



Vol. XXXVIII] SAVAGE: WARSZEWICZ AND GABB 279 

Other specimens collected by Warszewicz. 
At Krakow : 

Basciliscns mitratus. KM 932/1317 America. 
Stenostoma albifrons. KM 962/1296 America. 
Cyclophis aestivus. KM 981/1270 America. 
Pelamis bicolor. KM 989/1304 Pacific Sea. 
Lacerta miiralis viridis. KM 1019/1270 America. 
Bufo vulgaris. KM 1020 .America. 

Phyllomedma hypochondrica. KM 1024 1344 Guyana. 
Bujo chilensh. KM 1031/1349 Bolivia. 

At Berlin: 

*Rhiiiotyphlops albirostris. B 9529, 2 specimens; Veragua (Peters, 1857). 

*Anolis humilis. B 500; Veragua (Peters, 1863a). 

* Anolis intermedins. B 503; Veragua (Peters, 1863a). 

*Hyla sordida. B 3141; Veragua (Peters, 1863c). 

*Hyla piinctariola. B 4918; Veragua (Peters, 1863c). 

* St r atom ant is biporcatus. B 3222,3330; Veragua (Peters, 1863b). 

Bufo haematiticiis. B 3404; Veragua. 

Bufo typhonius. B 3442 ; Veragua. 

Most of the animals collected by Warszewicz are from what is today western 
Panama, but in his time constituted the Provincia de Veragua of the country of 
Nueva Granada (Colombia). Today the old Veragua comprises the Provincias 
de Veraguas, Chiriqui, and Bocas del Toro. In Warszewicz' day, western Veragua 
was called Chiriqui and the Atlantic lowlands were called Bocas del Toro. Several 
corrections seem necessary in dealing with the data associated with his specimens. 
First, all altitudes listed are extremely high and well above the known distribu- 
tions for the species. As I have previously pointed out (Savage, 1968) 19th 
Century Polish feet contained the equivalent of only nine English inches. There- 
fore I have given the corrected elevations in parentheses above. Kegel's (1867) 
report of Warszewicz' climbing 16,000 foot Volcan Chiriqui as previously cited 
shows the same point, since the mountain is 3478m. ( 1 1,31 1 feet) in height. Even 
these figures are out of the altitudinal range for several species, but since they 
were probably estimated, the differences are not extreme after the corrections 
have been made. 

Several forms described from Warszewicz' materials by Oskar Schmidt have 
never been retaken in Central America but, because of the lack of details regard- 
ing his route and the inaccessibility of the area on the continental divide visited 
by him, herpetologists have assumed that these animals would ultimately be 
rediscovered in the field. Recently, I (Savage, 1969) demonstrated that one 
species, Bujo vcraguensis, was based on a mislabeled Peruvian or Bolivian toad. 
At least three others, Hyla splendens, Hyla molitor, and Hyla moUtor marmorata 
may similarly be removed from any list of Central American amphibians. Savage 



280 



CALIFORNIA ACADEMY OF SCIENCES IProc. 4th Ser. 



83°00 



GABB'S ROUTE 



C. Utyum*' 



COSTA RICA 




10°00 



9°30 



9"00 



Figure 3. Map of southeastern Costa Rica, indicating area of Gabtj's collections and the 
route followed in his ascent of Cerro Utyum. 



Vol. XXXVIII] SAVAGE: WARSZEWICZ AND GABB 281 

and Heyer (1969) have shown that the latter two are totally unlike any Central 
American forms. Very likely they are also mislabeled South American frogs. 
The t5^e of Hyla splendens appears to be a female of the genus Gastrotheca. 
Charles F. Walker, the leading student of this genus, informs me that the type 
is very similar to some Peruvian Gastrotheca species and unlike any Panamanian 
or Colombian form. 

GABB AND THE EXPLORATION 
OF THE TALAMANCA 

William More Gabb was born in Philadelphia on January 20, 1839. He is the 
subject of a biographical memoir of the National Academy of Science (Dall, 1909) . 
Only details omitted from the memoir or matters directly related to his Central 
American experience are recounted here. Gabb was interested in geology and 
mineralogy and became associated with the California Geological Survey in 1862. 
As part of this work he spent the period 1862-1867 in California and was involved 
in the Survey's study of Baja California in the latter year. In 1869-1870 Gabb 
was active in geological work in Santa Domingo. 

Gabb came to Costa Rica in February, 1873, to undertake a study of the geog- 
raphy, geology, resources, and climate of the southeastern section of the country, 
the Talamanca (fig. 3). During the 19 months of his contract, 17 were spent in 
the field (until August, 1874). He returned to the United States in 1876 whence 
he again visited Santa Domingo. Malaria apparently contacted in Costa Rica 
was inflamed in Santa Domingo and he ultimately succumbed to tuberculosis of 
the lungs after his final return to the United States in April, 1878. He died in 
Philadelphia May 30, 1878. 

Gabb's fantastic activities during his Costa Rican stay are summarized in his 
reports (Gabb, 1875, 1877, 1913a, 1913b; Pittier, 1875, 1913). Most of his work 
was centered on the Valle de Talamanca, the region of the Rio Sixaola drainage. 
The upper portion of this area: The Valle de Rio Telire and the drainages of the 
Rio Uren, Rio Lari, Rio Coen, Rio Telire, and Rio Taberi, forms Alta Talamanca. 
The lower part of the valley from a line between Uatsi and the mouth of the Rio 
Yorkin to the coast is Baja Talamanca. The towering spires of the Cordillera de 
Talamanca border the Valle de Talamanca on the northwest. The Valle de Tala- 
manca was the original Spanish settlement in Costa Rica, where La Ciudad de 
Santiago de Talamanca was founded, near the present site of Suretka (60m.) on 
October 10, 1605. In Gabb's day the central settlement of the area (it had the 
only church and was the home of Mr. John H. Lyon, an American, who had 
administrative responsibility for the district) was San Bernardo de Sipurio ( 70m.) 
between the Rio Suedi and Rio L^ren about 3 miles above the mouth of the latter. 
This village and the Catholic church were destroyed by a flood of the Rio Lari in 
1909. A new mission was established at Amubri (75m.) in 1910 and serves as the 
central settlement in the area today. In 1873-74 about 1240 people lived in the 



282 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

region. Gabb was accompanied on most of his many trips through the area by 
two Costa Rican collectors, Jose Zeledon and Juan Cooper, both later famous 
naturalists in their own right, who collected most of the vertebrates. 

Gabb married an Indian girl, Victoria, and one son Guillermo was born to 
this marriage in 1874 or 1875. At least three grandchildren, Alfonso, Melania, 
and Francisco Gabb were alive in 1964 when I visited the Talamanca. Several 
great-grandchildren were also living including a Victoria Gabb, an exceedingly 
beautiful girl, who may have recalled the Indian beauty who married Gabb. 

Gabb visited almost every locality in the valley. As part of his fieldwork 
(Gabb, 1913b: 105-106, 114; 120-122, 127-128) he attempted to climb Pico 
Blanco (Cerro Kamuk) the highest peak (3554m.) in the southern Talamanca- 
Chiriqui range. Gabb tells it all — "We followed hunter's trails over a long, narrow, 
and very crooked ridge between the Uren and the Lari to a place called Bitsung- 
wo-ki, often scaling precipices, climbing around rocks, and in some parts scram- 
bling over bad places by means of ladders and bridges made of sticks placed 
there for this purpose. Beyond Bitsung-wo-ki, but two men had ever gone, and 
with one of them for a guide, we were forced to climb down to the Lari River, 
and ascend the mountains on the other side, to avoid impassable rocks. At the 
end of seven (7) days, we found ourselves on the side of a peak, which we as- 
cended, made our observations, and returned." His party consisted of 21 persons 
and subsisted mainly on pldtanos. They were on the peak June 13, 1873, after 
starting the ascent June 6. 

Gutierrez ( 1960) has conclusively shown that Gabb, by detouring up the Rio 
Lari, actually ascended Cerro Utyum (3084m.) (Cerro Cruz del Obispo) rather 
than Kamuk. The route followed by Gabb, his altitude record for the peak, 
9562 feet (2915m.), as well as his observations (1913b: 106) as pointed out by 
Gutierrez (1960) and confirmed by Carballo (1960) who scaled Kamuk, sub- 
stantiate this conclusion. Gabb apparently returned to Alta Talamanca via the 
Rio Lari. I ascended the latter river in 1964 to a point approximately where 
Gabb crossed over the ridge from the Rio Uren. This place is 3 days hard hiking 
from Amubri and lies at 800m., near the juncture of the Rio Dipari and Rio Lari. 

In view of these data, none of Gabb's animals should be listed from Pico Blanco 
(Cope 1875, 1876) but rather from Cerro Utyum. 

One of the principal supporters of Gabb's explorations was the legendary Costa 
Rican entrepreneur Minor C. Keith, then manager of what became the Costa 
Rican Northern Railroad, that today connects Puerto Limon and San Jose. 
Keith began the planting of bananas along the rail lines, originally to keep the 
railroad hands busy and to provide food. Gradually bananas became the basis 
for the development of the United Fruit Company. The Compafiia Bananera 
began to exploit the Valle de Talamanca in 1916. Poor and thin soils led to re- 
duction of activity in 1922. The Valle was abandoned to local farmers in 1925. 
A railroad that connected with the United Fruit Company lines in the Bocas del 



Vol. XXXVIII] SAVAGE: WARSZEWICZ AND GABB 283 

Toro region of Panama, at Guabito, formerly extended up river past Suretka. 
The bridge across the Rio Sixaola above Suretka was washed out in 1925 and the 
rails abandoned. In 1964 the railroad still ran from Sixaola to Volio (Uatsi). A 
truck road connects Puerto Viejo and Cahuita to Fields where another truck road 
runs to beyond Suretka. A jeep trail runs north from this road to Pandora in the 
Valle de Estrella. 

HERPETOLOGICAL SPECIMENS 
IN THE GABB COLLECTIONS 

The herpetological materials from Gabb's explorations were deposited at the 
United States National Museum (US) and reported on by E. D. Cope (1875, 
1876) in a large monograph. Many examples served as types of new taxa as 
indicated below by an asterisk (*). Cope's paper was originally published as a 
separate, with a limited letterpress run of 50 copies on November 26, 1876. The 
journal run (Cope, 1876) appeared early the next year. This original report on 
Gabb's material has been reissued as a special number of the journal O'Bios and 
may be purchased from the Departamento de Biologia, Ciudad Universitaria, 
Costa Rica. Because the Gabb material is well known I have indicated catalog 
numbers only for type materials. Unless otherwise denoted all specimens are from 
Provincia de Limon, Canton de Limon in Costa Rica. 

Specimens Collected by Gabb 

Siphonops mexicanus. Holotype, US 29762; Paratype US 29763; Limon (described as new 

species Siphonops proximus Cope, 1878) . 
Opheobatrachus vermicularis. One specimen from Cerro Utyum, 6000 feet (1830 m.) ; 2 ex- 
amples from lower country 20 miles (30 km.) from Coast. 
Oedipus mora ?. Eastern slope Cerro Utyum. 
*Cranopsis fastidiosns. Lectotype, US 32585; paratypes US 32584, 32586-87; Cerro Utyum, 

2500 feet (760m.). 
*Crepidins epiolicus. Cerro Utyum, 5000 feet (1520 m.) (type lost). Savage and Klugc, 1961. 
*Ollotis coerulescens. Cerro Utyum, 3000-5000 feet (915-1520 m.) (type lost). 
*Bnjo auritus. US 30676; east coast region (substitute name Bufo gabbi Taylor, 1952). 
Bufo valliceps. US 30592 ; eastern Costa Rica (described as new species, Bufo melanochloris 

Cope, 1878). 
Bufo agua. Eastern coast. 
Bufo haematiticus . Sipurio. 

Atelopus varius. Cerro Utyum and lower country. 

Dendrobates typographus. Low country about 10 miles (15 km.) inland. 
Dendrobates tinctorius. Lower country. 

"Dendrobates talamancae. Near Old Harbour on east coast (type lost). 
*Hyla gabbii. US 30658-59; near Sipurio. 
*Hyla uranochroa. US 30651; near Sipurio. 

*Hyla nigripes. US 30685-86; Cerro Utyum, 5000-7000 feet (1525-2135 m.). 
*Hyla elaeochroa. Lectotype, US 30689, paratypes US 30688, 30690; east foot of mountains 

near Sipurio. 
*Hyla punctariola pictipes. US 30652; Cerro Utyum, 5000-7000 feet (1525-2135 ra.). 



284 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

*Hyla punctariola monticola. US 30661, Cerro Utyum. 

*Phyllobates hylaejormis. US 30687; Cerro Utyum, 7000 feet (2135 m.). 

"Lithodytes podiciferus. US 30662, 30665-75 (US 30663 now at Harvard, US 30664 now at 

Michigan) ; Cerro Utyum, 5000-7000 feet (1525-2135 m.). 
* Lithodytes muricinus. Cerro Utyum (type lost). 
*Lithodytes habenatm. Cerro Utyum (type lost). 

* Lithodytes melanostictus. US 30608; Cerro Utyum, 7000 feet (2135 m.). 

* Lithodytes megalocephahis. US 32578; spur of Cerro Utyum, 6000 feet (1830 m.). 

* Lithodytes gulosus. US 32590; spur of Cerro Utyum, 6000 feet (1830 m.). 
*Hylodes cerasinus. US 32572 ; eastern slope of Cerro Utyum. 

Gnathophysia ocellaba. East side of the Cordillera. 

Ranula brevipalmata. Cerro Utyum. 

Mocoa assata. Old Harbour. 
*Mabuia alliacea. US 30619-20; from the low country. 

Mabuia cepedei. Below Sipurio. 

*Chalcidolepis metallicus. US 30568; Provincia de Alajucla, Fila de Aguacate. 
*Amiva gabbiana. US 32614-16; Old Harbour. 

Gerrhonotus fulvus. Summit of Cerro Utyum. 

Sphaerodactylus glaiicus. Near Sipurio. 

Thecadactylns rapicaudiis. North of Rio Estrella or North River. 

Anolis copei. Old Harbour (Puerto Viejo). 

Anolis trochilns. Talamanca. 

* Anolis pachypus. US 30683; slope of Cerro Utyum. 

* Anolis oxylophus. US 30556-57; Costa Rica. 
Anolis intermedins. 

Anolis capita. Old Harbour. 

Corythophanes cristatns. Sipurio. 

Iguana rhinophila. Low country. 

Basiliscus vittatus. Sipurio. 
*Basiliscus plumifrons. US 32622-6; Sipurio. 
* Xiphosoma anmdatum. US 32580. 

Boa imperator. Foot of mountains. 
* Leptognathus argus. US 30656; Sipurio. 
*Lcptognathus pictiventris. US 30657; eastern Costa Rica. 

Leptognathus nebidata. 

Sibon anmdatum. Old Harbour. 

Oxyrrhopus plumbeus. Low country 

Oxyirhopus petola. Sipurio. 
*Leptophis aeruginosus. US 30684; low country. 
*Leptophis saturatus. US 32563; Sipurio. 

Leptophis praestans. Sipurio. 
* Dendrophidium melanotropis. US 32597. 

Drymobius boddaertii. Talamanca. 

Herpetodryas carinatus. Low country. 

Spilotes corias. Talamanca. 
*Spilotes chrysobronchiis. US 30623; coast region. 

Coniophanes fissidens. Sipurio and Old Harbour. 

Rhadinaea decorata. Sipurio. 

Erythrolamprus vemistissimtis. Sipurio. 

Xenodon angustirostris. Sipurio. 



Vol. XXXVIII] SAVAGE: WARSZEWICZ AND GABB 285 

Stenorhina voitralis. Old Harbour. 
*Contia pachyura. US 30618; Sipurio. 
*Catastoma psephotum. US 62972; Cerro Utyum, 5000-7000 feet (1S2S-2135 m.). 

Elaps circinalis. Talamanca. 

Teleurapsis schlegelU. Eastern Costa Rica, Old Harbour to 5000-6000 feet (1525-1830 m.). 

Bothriechis nigroviridis. Cerro Utyum. 
'Bothriopsis proboscideus. Sipurio (type lost). 

Bothrops atro.x. Coast region. 
*Lachesis stenophrys. US 32479; Sipurio. 

Sphargis coriacea. Puerto Limon. 

Cinosternum leucostomum. Old Harbour and Sipurio. 
*Chelopus gabbii. US 45905. 
*Chelopus funerns. US 45900-01; 56134-35; Puerto Limon. 

Old Harbour, located on the coast between Punta Cahuita and the Boca de 
Sixaola, is now referred to as Puerto \'iejo de Limon. 

A FINAL WORD 

At certain levels both Warszewicz and Gabb were successful in their quest. 
The modern observer who has been over some of the same ground can only marvel 
at the courageous determination, dedication, and curiosity of these scientific 
pioneers. In regions sparsely settled, without roads or other communication, 
dominated by primitive and rugged terrain, demanding climate and debilitating 
disease, they still prevailed against all odds to open a new and exciting world to 
those that followed. They could do no more. 

ACKNOWLEDGMENTS 

My great thanks go to Dr. Henryk Szarski, Department of Comparative 
Anatomy, Jagellonian L'niversity, Krakow, for data and advice regarding Wars- 
zewicz and his career. He also supplied the accompanying photograph of the 
Polish botanist. Norman J. Scott, Department of Biology, University of Con- 
necticut, Storrs, Connecticut, was my field companion during an arduous follow- 
ing of Gabb's trails in the Talamancas and I appreciate his companionship and 
suggestions very much. Anthony J. Gaudin prepared the maps. This paper orig- 
inated while I was a John Simon Guggenheim Memorial Fellow 1963-1964 and 
I was encouraged in my studies in Costa Rica by the Organization for Tropical 
Studies. 

LITERATURE CITED 

Carb.\llo Gutierrez, Enio 

1960. Viaje a los vertices Dudu y Kamuk, en la Cordillera de Talamanca. Informe 
Semestral Instituto Geografico de Costa Rica, Enero-Junio: 1960, pp. 87-89. 
Cope, Edward D. 

1875. On the Batrachia and Reptilia of Costa Rica. Journal Academy of Natural 

Sciences, Philadelphia (letterpress), ser 2, no. 8, pp. 93-157. 

1876. On the Batrachia and Reptiha of Costa Rica. Journal Academy of Natural 

Sciences, Philadelphia, ser 2, no. 8, pp. 93-157. 



286 CALIFORNIA ACADEMY OF SCIENCES IProc. 4th Ser. 

1878. Tenth contribution to the herpetology of tropical America. Proceedings of the 

American Philosophical Society, vol. 17, no. 100, pp. 85-98. 
1893. Second addition to the knowledge of the Batrachia and Reptilia of Costa Rica. 

Proceedings of the American Philosophical Society, vol. 31, pp. 333-345. 
Dall, William H. 

1909. Biographical memoir of William More Gabb. 1839-1878. Biographical Memoirs 

National Academy of Science, vol. 6, pp. 347-361. 
Gabb, William M. 

1875. Informe sobre la exploracion de Talamanca verificada durante los Aiios de 1873^. 

Anales del Institute Fisico-Geografico Nacional de Costa Rica, vol. 5, pp. 67-92. 
1877. Aufnahme von Talamanca und der Kartographische Standpunkt von Costa Rica 

in 1877. Petermann's (Geographischer) Mittheilungen, vol. 23, pp. 385-387, 

pi. 18. 
1913a. Letter addressed by William M. Gabb to his Excellency General Don Tomas 

Guardia, President of the Republic of Costa Rica. Costa Rica-Panama Arbi- 
tration Documents, vol. 4, no. 581, pp. 95-97. 
1913b. Report of the Talamanca Exploration made during 1873 and 1874 by W. M. 

Gabb. Costa Rica-Panama .-Arbitration Documents, vol. 4, no. 582, pp. 97-142. 
Gutierrez Braun, Federico 

1960. Ascension de Gabb al Kamuk o Pico Blanco. Informe Semestral Instituto Geo- 

grafico de Costa Rica, Enero-Junio 1960, pp. 11-16. 
Peters, W. 

1857. Vier neue Amerikanische Schlagen aus der familie der Typhlopineri. Monats- 

bericht Koniglich Preussischen Akademie der Wissenschaften zu Berlin, 1857, 

pp. 402-403. 
1863a. Mittheilung uber eine neue Arten der Saurier-Gattung Anolis. Monatsbericht 

Koniglich Preussischen .'Akademie der Wissenschaften zu Berlin, 1863, pp. 135- 

149. 
1863b. Mittheilungen iiber eine neue Schlangen gattung, Styporhynchus, und ver- 

schiedene andere Amphibien des Zoologischen Museums. Monatsbericht Koniglich 

Preussischen Akademie der Wissenschaften zu Berlin, 1863, pp. 399-413. 
1863c. Mittheilungen Uber neue Batrachier. Monatsbericht Koniglich Preussischen 

Akademie der Wissenschaften zu Berlin, 1863, pp. 445-471. 
Pittier, Enrique 

1913. Introduction to the Spanish translation of the report of Mr. William M. Gabb, by 

Professor H. Pittier. Costa Rica-Panama Arbitration Documents, vol. 4, no. 

580, pp. 92-94. 
Regel, Eduard 

1867. J. V. Warszewicz. Gartenflora, Erlangen, vol. 16, pp. 95-96. 

ROUPPERT, KaZIMIEZ, ED. 

1927. J. Warszewicza. Ogroduictwo, vol. 23, pp. 1-32. 
Savage, Jay M. 

1968. The dendrobatid frogs of Central America. Copeia, 1968, no. 4, pp. 745-776. 

1969. Clarification of the status of the toad, Biifo veraguensis O. Schmidt, 1857. Copeia, 

1969, no. l,pp. 178-179. 
Savage, Jay M., and W. Ronald Heyer 

1969. The tree-frogs (Family Hylidae) of Costa Rica: diagnosis and distribution. 
Revista de Biologia Tropical, vol. 16, no. 1, pp. 1-12 7. 



Vol. XXXVIII] SAVAGE: WARSZEWICZ AND GABB 287 

Savage, Jay M., and Arnold G. Kluge 

1961. Rediscovery of the strange Costa Rica toad, Crepidius epioticus Cope. Revista 
de Biologia Tropical, vol. 9, no. 1, pp. 39-51. 
Schmidt, 0. 

1857. Diagnosen neuer Frosche des zoologischen Cabinets zu Krakau. Sitzungsberichten 

der Kaiserlichen Akademie der Wissenschaften in VVien, Mathematisch-Natur- 
wissenschaftlichen Klasse, vol. 24, no. 1, pp. 10-15. 

1858. Deliciae herpetologicae Musei Zoologica Cracoviensis. Beschreibung der im K. K. 

Museum zu Krakau befindlichen, von J. v. Warszewicz in Neu-Granada und 
Bolivia gesammelten Ungeschwanzten Batrachier. Denkschriften der Mathe- 
matisch-Xaturwissenschaftlichen Klasse der Kaiserlichen Akademie der Wissen- 
schaften in VVien, vol. 14, pp. 2.^7-258. 
Wagner, Moritz 

1853. Die Provinz Chiriqui (West-Veragua) in Mittel-Amerika. Petermann's (Geo- 
graphischer) Mittheilungen, vol. 13, pp. 16-24. 



PROCEEDINGS 

OF THE 

CALIFORNIA ACADEMY OF SCIENCES 

FOURTH SERIES 



Festschrift for George Sprague Myers 



Vol. XXXVIII, No. 15, pp. 289-298. December 31, 1970 



TELEOST HYBRIDIZATION STUDIES 

By 

Clark Hubbs 
The Vniversity of Texas at Austin 78712 

"Traditionally, studies such as ours have been based on morphology, espe- 
cially the skeleton, which is the only complete organ system available for detailed 
comparisons with fossils. However, with the variety of both primitive and ad- 
vanced teleosts living today, we are most emphatically of the opinion that ap- 
proaches other than morphological ones would be exceedingly fruitful in the in- 
vestigation of teleostean interrelationships." 

The above quotation from Greenwood et al. (1966) clearly states George S. 
Myers' philosophy that systematic studies are central to biology. Any difference 
or similarity between two groups of organisms can be of value in estimating the 
amount of divergence; therefore, all biologic investigations can provide direct 
or indirect taxonomic information. Similar or identical organisms should be used 
in order to obtain repeatable experimental results; therefore, all biological in- 
vestigations can be considered to be based on systematic research. 

At the present time, the classification of most major taxa is based on their 
gross anatomy. Although this is due primarily to tradition, there is a valid scien- 
tific basis. Most experimental* analyses cover such small fractions of the tax- 
onomic subdivisions that experiment-based classifications would have major 
gaps. INIoreover, a typological concept has no place in modern systematics. 



* Hereafter the word experimental should be considered to equal all types of analyses that are not tradi- 
tional studies of museum specimens. 

[289] 



290 CALIFORNIA ACADEMY OF SCIENCES fPROC. 4th Ser. 

When we contrast several related taxa we should compare the spectra of attri- 
butes in the diverse members of each taxon. Because of the difficulty of obtain- 
ing data and the short history of such studies, most experimental investigations 
contrast supposed typical representative members of the taxa to be compared 
( = typology). 

In contrast, experimental studies are not burdened with traditional taxonomic 
relationships. The refreshing new viewpoint can challenge the validity of an un- 
supported traditional taxonomic conclusion. The resulting interaction between 
morphological taxonomists and experimentalists can provide a realistic arrange- 
ment of organisms approximating their phylogenetic relationships. The primary 
contribution of the taxonomist to this interaction may be to point out taxonomic 
problem groups so that the experimental biologist can concentrate his efforts 
efficiently. 

Among the host of problems faced by the ichthyologist concerned with taxo- 
nomic problems has been the separation of environmental and genetic factors. 
Taning (1952) and many others have shown that a single environmental variable 
can concurrently alter several morphologic entities regardless of genotype. 
Therefore, a morphologic comparison that emphasizes those attributes might 
produce a dichotomy between those fishes reared in warm water and those reared 
in cold water, despite their genetic affinities. 

Environment-related problems can further plague taxonomy because survival 
in similar environments tends to select for similar morphologic attributes. For 
example, fishes that live in rock crevices tend to be elongate. Existence in this 
environment seems also to be enhanced by small eyes and scales. Some may be 
blind or naked. Many also have anteriorly located or reduced pelvic fins and 
small gill slits. Theoretically, distortions of true relationships by convergent or 
parallel evolution can be resolved by the use of diverse attributes. Practically, 
one must be careful not to use apparently divergent characters that happen to 
have selective value in similar habitats. Use of "non-adaptive" characters for 
taxonomy should help resolve problems of this type, but can one be certain that 
any character is not adaptive? 

Johnson and Wicks (1964) have advocated the use of molecular biologic 
( = electrophoretic) studies because they may provide the "ultimate" infor- 
mation on relationships. Studies of DNA hybridization seem to offer even more 
promise of approximating the degree of phylogenetic divergence. Even this "ul- 
timate" systematic tool may have potential weakness. Assume that we have two 
ancestral species of omnivorous fishes occupying estuaries. Both evolve into a 
freshwater herbivore and a saltwater carnivore. The DNA sequences in each 
species pair would diverge so that each saltwater type would have a sequence of 
codings favorable for survival in high salinity. Similarily each would have their 
DNA controlling their digestive enzymes designed to break down animal ma- 



Vol. XXXVIII] HUBBS: TELEOST HYBRIDIZATION 291 

terial. The freshwater representatives of each pair would have their nucleotide 
sequences designed to produce enzymes different from their sibling species, but 
the same as those of their more distantly related ecological counterpart. The 
above simplified model is undoubtedly extreme, but may indicate how depen- 
dence on a single analysis could be hazardous. It is also possible that this type 
of convergence would be missed despite the type of analysis used. 

In effect, we have returned to the premise that taxonomic conclusions should 
be based on the sum (or product?) of the biological studies available. Never- 
theless, each investigator should not attempt to carry out investigations in all 
areas, but should concentrate on those for which his aptitude, experience, and 
interest suit him. During recent years, I have used hybrid survival as an index 
of phylogenetic relationship. This approach has three major merits: 1 ) It neces- 
sitates minimal expenditures, 2) It measures genetic divergence, and 3) The 
results approximate those of classical morphological taxonomy. 

The general agreement between hybrid survival experiments and classical 
fish taxonomy (Hubbs, 1967) and the potential hazards in such tests (Hubbs 
and Drewry, 1960) combine to make such tests valuable contributions to, but 
not the ultimate answer for, problems of phylogenetic relationships. The exper- 
iments reported below relate to two levels of relationships: 1) Arrangement of 
fishes within the family Cyprinodontidae and 2 ) Arrangement of various fish 
families. 

MATERIALS AND METHODS 

The techniques of Strawn and Hubbs ( 1956) were used for removal and mix- 
ing of the gametes. Two modifications were used that reduced some of the ex- 
perimental difficulties. ]\Iost of our experiments have used gametes from "wild" 
fish, that is, the individuals were removed from natural populations when nearly 
ripe and taken to the laboratory for the experiments. This necessitated hurried 
field work to avoid having eggs shed or becoming overripe ( = stale) during 
transport. We have found that the gametes can be stripped and mixed in the 
field and then taken to the laboratory as they develop. Large numbers of ex- 
periments can be done in this manner if the trip is properly planned. We have 
used petri dishes for transportation of individual experiments. The ripe eggs 
attach to the surface of the basal unit and the top is held in place with rubber 
bands. A piece of tape on the edge of the basal unit permits water circulation, 
another on the bottom is used as a lable. The sets of petri dishes are placed in 
styrofoam containers and the water changed when necessary. One still must be 
careful not to remain away from the laboratory for too long because careful 
examination of development is difficult in the field and newly hatched larvae 
can escape from the petri dishes. Keeping the transportation equipment cool 
prolongs the time that one can remain in the field. 



292 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

We find that fertilization can be enhanced by use of mashed testes. Typi- 
cally, semen is removed from males by coelomic pressure. Some species have 
such small quantities of semen that milt is seldom extruded by this technique, 
but eggs can be fertilized by extracting and mashing the testes. The same tech- 
nique works with males whose sperm supply has been depleted in previous ex- 
periments. 

When pertinent, the sources of the stocks will be presented with the listing of 
the experiments. 

CROSSES OF CYPRINODONT FISHES 

Moenkhaus (1910), Newman (1908), Hubbs and Drewry (1960 and 1962), 
Archer (1966), Drewry (1967), and others have reported on many successful 
crosses among species of Fundulus and with species of related genera such as 
Adinia, Lucania, Rivulus, Jordanella, Crenichthys, and Cyprinodon. Unfor- 
tunately, the last two listed papers are in thesis form and have limited circulation. 
In general, the level of success parallels the estimate of phylogenetic similarity 
as determined by morphological taxonomy. Most species of Fundulus can be 
crossed with the others and the hybrids reared to mature size. Two previously 
unlisted crosses, F. seminolis 2 (Sumpter Lake, Florida) X F. cingulatus $ 
(Green Cove, Florida) and F. seminolis 9 (Sumpter Lake, Florida) X F. heter- 
oclitus $ (Matanzas Inlet, Florida) can be added to the extensive list of reared 
hybrids. 

Many authors have reported that hybrids between F. majalis or its near 
relative (race?) F. similis and other species of Fundulus will hatch if F. majalis 
type sperm is used and not if F. majalis type eggs are used. Six tests with F. 
zebrinus (Iraan, Texas) sperm and F. similis (9 mile pond, Texas) eggs had 
over 100 fertilized eggs fail to hatch. 

Drewry (1967) reported difficulties in crossing F. notatus or F. olivaceus 
with other species of Fundulus, but hybrids between them are easily reared 
(Thomerson, 1967). Drewry reported that only 1 of the 17 hybrids using F. 
olivaceus sperm (none available with F. notatus) hatched and it died shortly. 
The reciprocal experiments had 9 fertilized eggs of which 6 hatched but were not 
reared, indicating a difference between reciprocals that more recent results sup- 
port. Archer (1966) also failed to rear hybrids between Fundulus notatus or F. 
olivaceus and 2 other Fundulus species. Fundulus notatus 9 (Blanco R., Texas) 
has been crossed with F. kansae 6 (Colbert, Oklahoma) and F. grandis $ (Port 
Aransas, Texas) and one fish from 8 and 5 fertilized eggs respectively reared, but 
both were deformed. Fundulus olivaceus 9 (Scraper Park, Oklahoma) X F. 
kansae $ (Colbert, Oklahoma) had 13 of 18 eggs hatch but the deformed larvae 
shortly died. The difficulty of rearing the hybrids supports Drewry's observa- 
tions, but they can be reared. The reciprocal hybrids are much more difficult 



Vol. XXXVIII] HUBBS: TELEOST HYBRIDIZATION 293 

to rear. Fundulus kansae 2 (Colbert, Oklahoma) X F. olivaceus 6 (Scraper 
Park, Oklahoma) (twice); F. kansae ? (Colbert, Oklahoma) xF. notatus $ 
(Blanco R., Texas) (12 times); F. kansae 9 (Miller Cr., Texas) XF. nota- 
tus $ (Little Piney Cr., Texas), F. zcbrinus $ (Iraan, Texas) X F. notatus 6 
(Onion Cr., Texas) (5 times); F. cingulatus 2 (Dog Lake, Florida) X F. oli- 
vaceus $ (Baker, Florida), and F. similis 9 (Port Aransas, Texas) X F. nota- 
tus $ (Blanco River, Texas) all failed to develop late embryos. The "greatest 
level of success" achieved in these 22 tests was 1 egg {F. kansae Colbert X F. 
notatus Blanco) that produced chromatophores but did not show indication of 
gastrulation. The low success and high frequency of abnormalities supports 
Drewry's hypothesis that F. notatus and F. olivaceus are phylogenetically similar 
to each other but dissimilar to other species of Fundulus. Six of 7 attempts to 
cross F. notatus and F. olivaceus were successful but only 4 of 6 control tests. 

The previous intrageneric hybridization tests have not used the West Coast 
species, F. parvipinnis (Mission Bay, California, population). Failure of crosses 
with F. kansae (both reciprocals), F. grandis sperm, and eggs of F. cingulatus 
(3 tests, 2 populations), F. heteroclitus (2 tests), F. majalis, F. similis and F. 
olivaceus, indicates that this species is separate from other species now placed 
in Fundtdus. Of course, only 10 failures may not be enough tests to insure valid 
results. The better results of hybridizing F. parvipinnis with Crcnichthys baileyi 
(Hubbs, 1967) may indicate a common ancestry. 

Only Archer (1966) has previously reported hybridization tests with Jor- 
danella jloridae. He reported that one set of Jordanella eggs crossed with Cy- 
prinodon variegatus sperm died as late embryos. Three tests with Cyprinodon 
females from Iraan, Texas, resulted in no fertilization, but 1 of 2 and 3 of 4 
tests with female Fundulus zcbrinus and Lucania parva respectively from the 
same locality produced late embryos, but none hatched. This indicates that Jor- 
danella is distinct from those 3 species. Drewry (1967) reported difficulty in 
rearing hybrids between Lucania and Fundulus; however. Archer (1966) reared 
several F. pulvereus X L. parva individuals to adult size. Apparently the hybrids 
did not exhibit sexual dimorphism. 

INTRAFAMILIAL HYBRIDIZATION 

The relationship of hybridization success to phylogenetic divergence of tele- 
ost families may be shown to have significant importance in the taxonomic ar- 
rangement of living fishes. In part this approach resurrects those of Moenkhaus 
(1910) and Hertwig (1936). The ability of producing late hybrid embryos in a 
series of cyprinodontid X atherinid crosses has been considered support of their 
close relationship (Rosen, 1964). One more combination can be added to the 
series already reported (Hubbs, 1967, and citations). Two of 15 eggs from F. 
notatus 2 (Denison Dam, Texas) exposed to Menidia audens $ (U. Oklahoma 



294 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

Biological Station) sperm gastrulated. Both produced heads, one was attached 
to the yolk mass and the other was free. Black pigments covered the yolk mass 
on both. The one with the head free also had extensive orange pigmentation 
and remained alive until almost all of the yolk was expended. These results re- 
semble those of the reported series of cyprinodontid-atherinid hybrids. The fail- 
ure at gastrulation of a parallel experiment with Notcmigonus crysoleucas sperm 
shows that fundulines will not hybridize with all fishes. 

Menidia audens has not previously been reported to have been tested in 
rearing experiments. The series (4) of controls done the same day all had 50' 
percent plus fertilization and hatching. Similar to most atherinids, the controls 
died a week later apparently due to starvation. Intrafamilial hybrids between 
M. audens 9 and Labidesthes sicculus <5 (Tishomingo, Oklahoma) (twice) had 
25 percent-50 percent fertilization and all hatched and died with the maternal 
controls, showing that Menidia hybrids can be reared as far as the controls. 

Menidia audens has also been tested for hybridization survival with members 
of several other families. The tests with Notropis cornutus, N otemigonus cryso- 
leucas (3 times), Gasterosteus aculeatus, and Aphredoderus sayanus all termi- 
nated before gastrulation. Gastrulation and embryonic formation occurred in 
tests with a centrarchid and several percids. One egg gastrulated among 2 sets of 
M. audens eggs exposed to Lepomis macrochirns sperm. It died before pigmenta- 
tion. Another set of M. audens eggs exposed to Etheostoma radiosum (Blue 
River, Oklahoma) sperm produced 5 early embryos, of which only 2 developed 
pigmentation, the test with Percina caprodes males from the same locality failed. 
Seven reciprocal experiments were set up with percid eggs from Blue River fe- 
males, one E. spectabile and six E. radiosum. Six had some development and 58 
of 192 gastrulated embryos produced pigmentation, but none showed any sign of 
circulatory development. Clearly these hybrids were more successful than most 
other interfamilial tests and approached that of the atherinid-cyprinodont tests. 

A series of other intrafamilial tests were done with Etheostoma or Hadrop- 
terus eggs. It is not surprising that all 7 tests with the ostariophysines, Moxo- 
stoma poecilurum, Notropis cornutus, Notropis umbratilis, Notcmigonus cry- 
soleucas, and Opsopoedeus emiliae failed to gastrulate or that 3 crosses with 
the "black race" of Gasterosteus aculeatus (Chehalis, Washington) did not gas- 
trulate. Five of 17 tests with Aphredoderus sayanus sperm had 1 or more eggs 
gastrulate. The males were from a stock obtained at Douglass, Texas. The suc- 
cessful combinations were essentially the same ones as the failures, indicating 
that the data are representative of the interfamilial combination. Two sets of E. 
asprigene (Douglass, Texas) eggs each had 2 eggs gastrulate and develop eye 
pigmentation. Two of them formed a heart that beat irregularly, but had no vis- 
ible cells in the tubes. Many erythrocytes were present on the yolk mass anterior 
to the head. Another embryo had no heart beat but died breaking the egg shell. 



Vol. XXXVIII] HUBBS: TELEOST HYBRIDIZATION 295 

Two sets of H. scicrus (San Marcos and Pedro Creek, Texas) had 2 and 3 eggs 
gastrulate. Only 3 developed recognizable heads. Three additional embryos 
from an E. spectabile (San Marcos, Texas) X A. sayanm cross that developed 
pigmented eyes were sacrificed in an unsuccessful attempt to analyze the cy- 
tology. The putative hybrids were diploid but chromosome markers were not 
noted. 

Four of 1 1 sets of Etheostoma eggs exposed to Elassoma zonatum sperm had 
gastrulation. All but one involved E. spectabile eggs. The 2 successful sets from 
Shoal Creek, Missouri, females had 6 gastrulated eggs, 5 had eye pigmentation, 
and 1 a functional heart beat and flow. All died at the time the maternal con- 
trols hatched. The 2 successful tests with Blue River, Oklahoma, females {E. 
radiosum and E. spectabile), each had 1 embryo that developed eye pigmen- 
tation. 

The failure of 2 sets of G. aculeatus eggs exposed to A. sayanus sperm indi- 
cates that A. sayanus sperm will not fertilize all teleost eggs. 

DISCUSSION 

Most of the intrafamilial hybrid experiments substantiate those previously 
reported, i.e., teleost hybrids are relatively easily produced and if the parental 
morphology is similar the hybrids are easily reared. Drewry had evidence of a 
genetic block to the development of hybrids between F. notatus or F. oUvaceus 
and other members of that genus. Because he had few tests (26 fertilized eggs), 
it might be possible that his results were due to chance. We had 26 fertilized 
eggs in the least studied reciprocal {F. olivacetis or F. notatus egg) and 22 tests 
with the reciprocal. Not only did the results confirm those of Drewry, they also 
showed a distinct difference between the reciprocals. Because most of our 
"standard Fundulus" in these tests were of the F. kansae — F. zebrinus type, it 
may be that the difference between reciprocals may relate only to those combi- 
nations. It is possible, however, that we have a second difference in reciprocal 
hybrid survival in funduline fishes. It is amply evident, however, that F. notatus 
and F. olivaceus are quite distinct from other species of Fundulus. Hybridization 
tests have clearly showed that gametes from Adinia xenica and Lucania parva 
are more compatible with gametes from typical Fundulus than are those of Fun- 
dulus notatus or F. oUvaceus. Therefore, the genus should be expanded to in- 
clude the members of Lucania and Adinia, or F. notatus and F. olivaceus should 
be separated from the other Fundulus species and placed in the genus Zygonectes. 

The failure of 10 tests between Fundulus parvipinnis and 7 other species of 
Fundulus indicates that this species is quite distinct. The reasonable success of 
hybrids between F. parvipinnis and the Crenichthys-Empetrichthys complex 
suggests a possible relationship that makes biogeographic sense. The hybrids of 
all west coast Fundulines have earlier developmental blocks when crossed with an 



296 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

east-coast representative than when crossed with another western type. Perhaps 
the Fundulines had an east-west primary separation. The western types then 
diverged into a coastal {F. parvipinnis precursor) and a desert spring (Ci'e- 
nichthys, Empetrichthys) type. Because F. parvipinnis occupied a "typical 
Fundulus" environment, it retained "typical Fundulus" morphology, while Cre- 
nichthys and Empetrichthys occupied clear warm spring waters and became su- 
perficially distinct. 

Assuming that the cyprinodonts have lost spinous fin rays, it seems that hy- 
brids between soft-rayed and spiny-rayed fishes die at or before gastrulation. 
The consistent failures indicate a uniform distant relationship and perhaps a 
treelike phylogeny rather than a bushy type suggested by Greenwood et al. A 
treeUke phylogeny is also indicated by hybrid development to late embryonic 
stages in almost all crosses between two spiny rayed types. If we placed Aphredo- 
derus and Etheostoma in separate superorders than the hybrid success is incon- 
gruous; in contrast, if we follow Regan and make Aphredoderus a relic of a group 
ancestral to other perciforms, the phylogeny would agree with the hybridization 
data. 

The success of most interfamilial spiny rayed hybrids is quite similar. Only 
two types of combinations usually survive to hatching. The cyjorinodont- 
atherinid hybrids and hybrids among percids, serranids, and centrarchids. Re- 
cent classifications place each unit in a suborder. 

In the future, hybrid survival tests may aid in determining other family 
group relationships. 

ACKNOWLEDGMENTS 

The experiments reported in this paper were supported by NSF GB 6429. 
Many graduate students at the University of Texas at Austin aided in the col- 
lection of parental stocks. The stocks of Fundulus parvipinnis were obtained 
through the courtesy of Dr. Carl L. Hubbs at the University of California at 
San Diego. 

This paper is dedicated to Dr. George S. Myers because of his inspired teach- 
ing of systematic ichthyology. 

LITERATURE CITED 

Archer, James Douglas 

1966. The behavior and characteristics of some hybrids of cyprinodont fishes (Atherini- 

formes). Unpublished M.S. Thesis, Cornell University, pp. i-vii + 1-70. 
Drewry, George Earl 

1967. Studies of relationships within the family Cyprinodontidae. Unpublished Ph.D. 

Dissertation, The University of Texas, pp. i-vii -f 1-134, text figs. 1^. 
Greenwood, P. Humphrey, Donn E. Rosen, Stanley H. Weitzman, .\nd George S. Myers 
1966. Phyletic studies of teleostean fishes, with a provisional classification of living 
forms. Bulletin of the American Museum of Natural History, vol. 131, pp. 339- 
456, text figures 1-9, plates 21-22, charts 1-32. 



Vol. XXXVIII] HUBBS: TELEOST HYBRIDIZATION 297 

Hertwig, Paula 

1936. Artbastarde bei Tieren. Handbuch Verebungswissenschaft, vol. 2, pp. 1-140. 
HuBBS, Clark 

1967. Analysis of phylogenetic relationships using hybridization techniques. Bulletin 
of the National Institute of Sciences of India, No. 34, pp. 48-59. 
HuBBS, Clark, and George E. Drewry 

1960. Survival of Fi hybrids between cyprinodont fishes, with a discussion of the cor- 
relation between hybridization and phylogenetic relationship. Bulletin of the 
Institute of Marine Science, The University of Texas, vol. 6, pp. 81-91. 
1962. Artificial hybridization of Crenichthys baileyi with related cyprinodont fishes. 
Texas Journal of Science, vol. 14, pp. 107-110. 
Johnson, Murray L., and Merrill Wicks 

1964. Serum-protein electrophoresis in mammals: significance in the higher taxonomic 
categories. In: Leone, Charles A., Taxonomic biochemistry and serology, 
Ronald Press, New York, pp. 681-700, 7 text figures. 
Moenkhaus, J. 

1910. Cross fertilization among fishes. Proceedings of the Indiana Academy of Science, 
for 1910, pp. 353-393. 
Newman, H. H. 

1908. The process of heredity as exhibited by the development of Fundulus hybrids. 
Journal of Experimental Zoology, vol. 5, pp. 503-561. 
Rosen, Donn Eric 

1964. The relationships and taxonomic position of the halfbeaks, killifishes, silversides, 
and their relatives. Bulletin of the American Museum of Natural History, vol. 
127, pp. 219-267. 
Strawn, Kirk, and Clark Hubbs 

1956. Observations on stripping small fishes for experimental purposes. Copeia, 1956, 
pp. 114-116. 
Taning, a Vedel 

1952. Experimental study of meristic characters in fish. Biological Reviews, vol. 2 7, 
pp. 169-193, text figures 1-10. 
Thomerson, Jamie E. 

1967. Hybrids between the cyprinodontid fishes, Fundulus notatus and Fundulus oli- 
vaceus, in southern Illinois. Transactions of the Illinois Academy of Science, 
vol. 60, pp. 375-379. 



PROCEEDINGS 

OF THE 

CALIFORNIA ACADEMY OF SCIENCES 

FOURTH SERIES 




Festschrift for George Sprague Myers 



Vol. XXXVIII, No. 16, pp. 299-340; 5 figs.; 4 tables. December 31, 1970 



A REVISION OF THE FISHES OF THE GENUS 

T^OTOTHENIA 

FROM THE NEW ZEALAND REGION, 

INCLUDING MACOUARIE ISLAND 

By 

Hugh H. DeWitt 
Marine Science Institute, University of South Florida, St. Petersburg, Florida 33701 

INTRODUCTION 
While preparing a revision of the southern and Antarctic fishes of the genus 
Notothenia, it became evident that the taxonomy of the species found in the 
New Zealand region is confused. In the most recent review (Parrott, 1958), 
five species are identified as occurring there. Of these, only four are valid, only 
three of the four are found in the New Zealand region, and the nomenclature of 
the three is entirely confused. This should not reflect upon Parrott, for he fol- 
lowed Boulenger, Waite, Regan, and Norman. For these reasons it is timely to 
present new descriptions and a new key for the New Zealand species together with 
a clarification of the nomenclatural confusion which has surrounded them. 

I include Macquarie Island in this paper because two of the three species of 
Notothenia recorded from there also occur in New Zealand waters. Further, 
Notothenia coriiceps is included in the key to the species because it is widely dis- 
tributed in the Southern Ocean, is known from the Kerguelen Islands, and even- 
tually also may be found at Macquarie Island; a description of it is not given. 



^Contribution number 15 from the Marine Science Institute of the University of South Florida. 

[299] 



300 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

I have not included N otothcnia cornucola in this revision because I do not 
believe it occurs at New Zealand. This species was early recorded from New 
Zealand waters (Giinther, 1860; p. 262; Hutton, 1872; p. 26; 1873; p. 262) 
and continues to be included in lists of New Zealand fishes although no specimens 
identified as A^. cornucola have been found for nearly 100 years. The most re- 
cent reference is Parrott (1958), who admits that its occurrence is doubtful, 
although he includes it in his key to the New Zealand species of Notothenia. 
Considering that specimens of A", cornucola are encountered most commonly in 
littoral and shallow inner sublittoral areas (for example, among and under the 
rocks of the beaches near Punta Arenas, Chile, at low tide), it seems likely that 
if the species actually occurred in the New Zealand region it would be well 
known there. Norman (1937b; p. 86) reviewed the evidence and concluded that 
it ". . .is very slender." The specimen recorded by Giinther was probably mis- 
labeled, and Mutton's 1872 record is probably based upon that of Giinther. 
Hutton's 1873 record from the Chatham Islands was probably based upon 
specimens of A^. angustata. This species has recently been collected there 
(Moreland, 1957) and I have seen the specimens (see the section on material 
examined under A^. angustata). Further, Hutton's (1873) statement that the 
upper lateral line "... extends to the end of the second dorsal, . . ." agrees best 
with my observations of A^. angustata rather than with iV. magellanica, the 
species to which Norman believed Hutton referred. For these reasons I have 
included Hutton's 1873 reference to A^. cornucola in the synonymy of A'^. an- 
gustata. I have not included the listings of A^. cornucola Richardson found in 
the lists and catalogues of New Zealand fishes because, in that form, they refer 
to a species which I believe does not occur in New Zealand. 

Several check lists of New Zealand fishes have been prepared at various 
times, some of which I have not seen. The most important are those by Gill 
(1893), which reviews in detail the earlier works, and by Phillipps (1927b) 
which refers to earlier lists. For lists that I have seen, I have included the 
references to Notothenia. species in the synonymies according to my present 
interpretations of the names used. For example, the name Notothenia microlepi- 
dota is listed under that species even though during that period the name was 
used in reports on collections for specimens properly called A^. angustata. 

MUSEUM ABBREVIATIONS 

In preparing my descriptions I have utilized specimens from the collections 
of museums whose names are abbreviated in the lists of material examined as 
follows. 

BMNH: British Museum (Natural History), London. 

CM: Canterbury Museum, Christchurch, New Zealand. 

DM: Dominion Museum, Wellington, New Zealand. 

MACN: Museo Argentino de Ciencias Naturales, Buenos Aires. 



Vol. XXXVIII] DeWITT: NOTOTHEXIA FROM NEW ZEALAND 301 

MLP: Museo de La Plata, La Plata, Argentina. 

NMV: Naturhistorisches Museum, Vienna. 

PM: Museum National d'Histoire Naturelle, Paris. 

SAM: South Australian Museum, Adelaide. 

SU: Division of Systematic Biology, Stanford University, Stanford, Cali- 
fornia. 

VSC-Eltanin: material collected by the University of Southern California 
Antarctic Biological Research Program from the USNS Eltanin. 

USNM: United States National Museum, Washington, D.C. 

ZIL: Zoological Institute, Leningrad. 

ZMB: Zoologisches Museum, Humboldt- Universitat, Berlin. 

MEASUREMENTS AND COUNTS 

All measurements were made in a straight line with calipers, and are pre- 
sented in the descriptions as thousandths of the standard length unless otherwise 
specified. All were made on the left side unless there was a deformity or loss 
which necessitated using the right side. Lateral line and pectoral fin counts 
were usually made on both sides. Those measurements which are not usually 
made, or which have been made differently in the past, are defined in the 
following alphabetical list. 

Anal to Pelvic Distance: from base of pelvic spine to origin of anal fin. 

Body, Depth of: measured at origin of anal fin. 

Body, Width of: measured at thickest part of body above origin of anal fin. 

Dorsal Interspace: distance between base of last spine of first dorsal fin and 
first ray of second dorsal fin. 

Dorsal to Anal Distance: distance between origins of second dorsal and 
anal fins. 

Dorsal to Caudal Distance: distance between last ray of second dorsal fin 
and midbase of caudal fin. 

Head. Depth of: measured at vertical through cheeks. 

Head, Length of: measured from tip of snout (upper jaw) to posteriormost 
edge of opercular flap. 

Head, Width of: distance between cheeks. 

Pectoral Fin, Length of: measured from base of uppermost ray to tip of 
posteriormost extending ray. 

Pectoral to Pectoral Distance: distance between upper ends of bases of 
pectoral fins. 

Pelvic Fin, Length of: measured from base of pelvic spine to tip of posterior- 
most extending ray. 

Post Orbital Distance (Postorbital Part of Head): measured from posterior 
margin of orbit to posteriormost edge of opercular flap. 

Standard Length: measured from tip of upper lip to midbase of caudal fin. 



302 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

Upper Jaw, Length of: measured from tip of upper lip to posterior end of 
maxillary. 

The counts for the caudal fin include all the branched rays plus 1 additional 
ray above and below, /. e., the branched rays plus 2. The last ray elements in 
the second dorsal and anal fins are counted separately. The scales in a lateral 
longitudinal series are counted from the upper end of the base of the pectoral 
fin to the base of the caudal fin. Gill raker counts are given as follows: 
6-9 + 0-1 + 12-17 = 18-26. This means that there are a total of 18-26 gill 
rakers, of which 6 9 are on the upper limb, none or 1 at the angle, and 12-17 
on the lower limb. On each arch, except occasionally the fourth arch, there are 
2 rows of gill rakers, one projecting anteriorly and the other posteriorly. These 
are called, respectively, the anterior and posterior series. The lateral lines and 
their counts as well as the terminology of the cephalic canals have already been 
described (DeWitt, 1962). 

GENUS NOTOTHENIA RICHARDSON 

A formal diagnosis of the genus will be presented elsewhere. The following 
characters serve to distinguish it from other genera of New Zealand marine 
fishes. The nostrils are tubular and single on each side; the New Zealand 
species have the hind margin of the tube extended into a flap. The gill mem- 
branes are joined to each other and to the isthmus, forming a free fold across the 
isthmus. The vomer and palatines are edentulous. Two dorsal fins are present, 
the first composed of 3-8 spines which are usually soft and flexible, the second 
long and composed of soft rays. The anal fin is similar to the soft dorsal fin. The 
pectoral fins have broad, almost vertical, slightly curved bases. The body is 
scaled; the head is nearly naked in the New Zealand species. The scales may 
be ctenoid or nonctenoid, with both types usually present. Two lateral lines are 
present on the body, one high near the bases of the dorsal fins, the other on the 
midside in the region of the caudal peduncle. In the New Zealand species the 
head is somewhat depressed, and the interorbital space and the top of the head 
are broad and flat. 

KEY TO THE SPECIES 

la. Lateral scales 78-99; middle lateral line 24-37; upper lateral line 61-75; 15-19 gill 
rakers on lower limb of first gill arch; total number of gill rakers on first arch 
24-30 N. microlepidota, p. 325. 

lb. Lateral scales 73 or less; middle lateral line 23 or less; upper lateral line 30-61; 
8-15 gill rakers on lower limb of first gill arch; total number of gill rakers on first 
arch 15-23 - - 2. 

2a. (from lb). Pectoral rays 21-24 - N. rossii, p. 312. 

2b. Pectoral rays 16-19 3. 

3a. (from 2b). Second dorsal fin with 35-41 rays; anal fin with 26-32 rays .— N. coriiceps. 

3b. Second dorsal fin with 27-31 rays; anal fin with 22-26 rays 4. 



Vol. XXXVIII] DeWITT: NOTOTHENIA FROM NEW ZEALAND 303 

4a. (from 3b). Upper lateral line with 36-48 tubular scales; total number of scales in 
upper and middle lateral lines 45-57; length of caudal peduncle 37.0-45.5 percent of 
head length; preoperculo-mandibular canal not connected with the temporal canal; 
dorsal surface of head without prominent ridges N. magellanica, p. 303. 

4b. Upper lateral Hne with 45-61 tubular scales; total number of scales in upper and 
middle lateral lines 59-76; length of caudal peduncle 25.5-34.5 percent of head 
length ; preoperculo-mandibular canal connected dorsally with the temporal canal ; 
in larger specimens prominent ridges present on top of head extending from above 
each eye posteriorly onto temporal region N. angustata, p. 318. 

Notothenia magellanica (Forster). 

Gadus magellanicus Forster, in Bloch and Schneider, 1801: 10-11 (original description; 
type locality seas about Tierra del Fuego; no types preserved, description based upon 
notes taken from fresh specimens and an unpublished rough drawing) ; Forster, 1844: 
361-362 (description) ; Richardson, 1846: 61 (listed in footnote, see under Lota 
magellanica, below) . 

Notothenia magellanica Richardson, 1844: 9 (counts with reference to illustration: "Icon, 
ined. Bibl. Banks, fig. 178," catalogued in British Museum (Natural History) in 
Banksian MSS. no. 6 & 7) ; Gill, 1862: 520 (listed). 

Lota magellanica Richardson, 1846: 61 (possibly a mistaken generic assignation"); Gill, 
1862: 520 (listed). 

N otothenia magellanicus Gunther, 1860: 260 (listed). 

Notothenia magellanicuz Delfin, 1899a: 21 (listed). 

Notothenia macrocephalus Gunther, 1860: 263 (original description; type locality Falkland 
Islands; type in British Museum); Gill, 1862: 520 (listed); Cunnlngham, 1871: 470 
(color notes); Perugl\, 1891: 618-619 (description); Smitt, 1897; 9-12, pi. 3, figs. 
23-26 (description, scales) ; Boulenger, 1900: 53 (listed). 

Notothenia maoriensis Haast, 1873: 276, pi. 16 (original description; type locality near 
Lyttleton Harbour, New Zealand; present location of type unknown, probably lost); 
Hutton, 1876: 212-213 (description); Hutton, 1890: 279 (listed); Gill, 1893: 118 
(listed); Waite, 1907: 29 (listed); Frost, 1928: legend for pi. 17, fig. 15 (otolith). 

Notothenia antarctica Peters, 1876: 837 (original description; type locality Accessible Bay, 
Kerguelen Island; type in Zoologisches Museum, Humboldt-Universitat, Berlin). 

Notothenia antarcticiis Studer, 1879: 131 (listed; color notes). 

Notothenia hassleriana Steindachner, 1876: 69-70, pi. 6, left-hand figures (original descrip- 
tion; type localities Puerto Bueno and Port- Gallant, both in Strait of Magellan; types 
in Naturhistorisches Museum, Vienna); Steindachner, 1898: 303 (listed). 

Notothenia argiita Hutton, 1879: 339 (original description; type locality Campbell Island; 
type in British Museum) ; Hutton, 1890: 280 (listed) ; Gill, 1893: 118 (listed) ; Waite, 
1907: 30 (listed). 

Notothenia niacrocephala Gunther, 1881: 20 (listed); Vaillant, 1888: 27, pi. 3, figs. 2a-d 
(listed, illustrations); Boulenger, 1902: 186 (listed); Steindachner, 1903: 207 (listed); 
Dollo, 1904: 86 (Hsted, distribution); Lonnberg, 1907: 10 (listed, color notes); Regan, 
1913: 277 (description, distribution); Hussakof, 1914: 89 (listed with counts); Waite, 



- In his description of Lota breviuscula. Richardson compares L. bieviuscula with several other species, 
among which is "Lota magellanica of Forster." In a footnote he lists the species and gives some data for 
each. Here Forster's species is listed as Gadus magellanicus, with the following counts: B. 6; D. 5-31; 
A. 25; C. 14; P. 17; V. 6. These counts are identical with those given in Forster (in Bloch and Schneider, 
1801: 11; 1844: 362) and Richardson (1844: 9) under Gadus magellanicus. except that Richardson does 
not give an anal fin count. It seems obvious that both Lota magellanica and Gadus magellanicus refer to 
the same fish, but the reason for the use of Lota is unclear to me. 



304 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

1916: 66-69, pi. 3, fig. 2 (description, illustration) ; Thompson, 1916: 431-433 (descrip- 
tion) ; Regan, 1916: 378-379 (distribution); Phh-lipps, 1921: 123 (listed); Thompson 
and Anderton, 1921: 94 (listed, synonymy); Rendahx, 1925: 6 (listed); Phillipps, 
1927a: 13 (listed); Phillipps, 1927b: 44 (listed); Frost, 1928: 454-455, pi. 17, fig. IS 
(otolith) ; Norman, 1937b: 88-90 (description, illustration, distribution) ; Norman, 1938: 
27 (distribution); Oliver Schneider, 1943: 110 (listed, illustration); MacDonagh and 
Covas, 1944: 235-236 (description, distribution); Fowler, 1945: 128-129 (listed); 
Hart, 1946: 339 (pelagic young) ; Fowler, 1951: 314 (key) ; Andriashev and Tokarev, 
1958: 199 (listed); Andriashev, 1959: 5 (vertebral count); Blanc, 1961: 124 (descrip- 
tion); Kenny and Haysom, 1962: 252 (habitat, food); Slack-Smith, 1962: 14 (color 
notes, habitat, food). 

Material examined. USNM 77329: Sandy Point (Punta Arenas), Strait of 

Magellan, 53°10'S., 70°SS'W. (1; 183 mm.). 
USNM 88755: Municipal jetty (Port Stanley?), Falkland Islands (1; 

193 mm.). 
USNM 88756: Mullet Creek, Falkland Islands, 51°44'S., 57°53'W. (2; 51.9 

and 55.3 mm.). 
USNM 171000: Kainan Bay, Ross Sea, Antarctica, 78°14'S., 16r5S'W. 

(1; 229 mm.). 
SU 59880: Macquarie Island (3; 48.0-169 mm.). 
SU 59882: Macquarie Island (2; 139 and 168 mm.). 

BMNH 1860.2.20.2: Falkland Islands (1, a skin; holotype of N. macrocephda) . 
BMNH 1886.11.18.28: Campbell Island, from Otago Museum, Dunedin, New 

Zealand ( 1 ; 150 mm. ; type of A", arguta) . 
ZMB 21626: Deutsche Tiefsee-Expedition Station 123, 49°07'S., 08°40'E.; 

bottom depth 4418 m.; presumably taken at surface in a plankton net, 22 

November, 1898 (1; 80.2 mm.). 
NMV 59926: Port Gallant (Puerto Gallant), 53°40'S., 71°58'W., field no. 

1203a (1; 86.0 mm.; lectotype of N. hassleriana) . 
NMV 65389: Puerto Bueno, 50°59'S., 74°12'W., field no. 1203b (1; 87.0 mm.; 

paralectotype of A^. hassleriana) . 
MACN 1859: Punta Colnet (Cabo Colnett, 54°43'S., 64°20'W.), 17 fathoms 

(1; standard length not measured). 
MACN 2673a: Bahia Tethis (Tierradel Fuego), (1; 155 mm.). 
ZIL (no number): Transvaal Cove, Marion Island, about 2 meters (2; 189 

and 216 mm.). 
ZIL (no number) : Scotia Sea, 60°38'S., 44°08'W., bottom depth 287 m.; depth 

of capture 0-60 m.; gear Isaacs Kidd trawl; at Academician Knipowich 

Station 85 (1; 261 mm.). 
CM (no number): South Island, New Zealand, probably near Dunedin (1; 

137 mm.). 

I have also examined specimens deposited in New Zealand museums (all 



Vol. XXXVIII] DeWITT: NOTOTHENIA FROM NEW ZEALAND 305 

uncatalogued) from the following localities. DM: Campbell Island, from Camp 
and Garden coves after tidal wave. CM: Campbell Island; Tucker Cove, 
Campbell Island; Penguin Harbour, Campbell Island; Perseverance Harbour, 
Campbell Island; Maccjuarie Island, 17 fathoms. 

Description. Body evenly curved both dorsally and ventrally from head to 
base of caudal fin; compressed posteriorly, becoming broader and more rounded 
toward head; greatest depth of body at about origin of second dorsal fin; depth 
of body 208-282, its width 122 150; pectoral to pectoral distance 144-225; 
dorsal to anal distance 237-306. Caudal peduncle longer than deep, its length 
107-135, its depth 93-102; dorsal to caudal distance 104-138. Head slightly 
shorter than average for genus, its length 280-320; its width, 146-248, about 
equal to its depth, 198-224. Vertebrae 16-18 + 28-30 = 45-47. 

Snout very bluntly rounded from dorsal view; from lateral view it rises 
steeply from tip of upper jaw to a point a little above and anterior to nostrils, 
where it becomes abruptly less steep; its length 82-102. Tubes of nostrils short, 
with posterior rim raised into a flap which may be folded over opening; placed 
52-79 from tip of snout, 17-29 from orbit and 52-75 apart. Eyes placed high 
on sides of head, but below dorsal profile; diameter of orbit 58-96. Interorbital 
region very broad and flat, its least width 88-134; all of top of head, from pos- 
terior part of snout to occipital region, nearly straight and rising slightly pos- 
teriorly; length of postorbital part of head 141-176. 

Jaws short but wide, maxillary extending posteriorly to about vertical from 
pupil of eye; length of upper jaw 94-115. Teeth in each jaw in two almost uni- 
serial bands; those in outer bands much larger and more numerous than those 
of inner bands and extend full length of jaws; inner bands confined to anterior 
% or less of jaws. The numbers of teeth vary, for in some individuals the bands 
are almost entirely uniserial, whereas in others they may become essentially 
double for part of their length. Oral valves extend most of length of each jaw, 
the lower broadest; their exposed surfaces covered with coarse papillae, espe- 
cially close behind inner bands of teeth. Tongue fleshy and densely covered 
with short, slender papillae which may be covered by a mucous coating and 
appear as low rounded papillae. 

Anterior gill rakers of first gill arch nondentigerous, or occasionally with 1 
to a few spines, the larger ones flattened, arranged 3-6 + 1 + 9-13 = 14-19. 
Posterior gill rakers of first arch dentigerous, arranged 0-1 + 0-1 + 10-15 = 
12 - 16. Gill rakers of remaining arches all dentigerous; 1-11 in posterior series 
of fourth arch. Branchiostegal rays 6; pseudobranchiae curved ventralward 
posteriorly. 

First dorsal fin 3-6, originating 306-343 from tip of snout, from just behind 
to just in advance of upper end of base of pectoral fin; lower than second 
dorsal fin, second or third spine longest, 67-99. Second dorsal fin 29 31, origi- 



306 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

nating 396-437 from tip of snout, 25-65 behind base of last spine of first dorsal 
fin; length of sixth ray 125-171, of sixth from last ray 87-105. Membrane be- 
hind last spine of first dorsal fin may reach to base of first ray of second dorsal 
fin. Anal fin 22-26, originating 513-606 from tip of snout, below bases of rays 
8-10 of second dorsal fin; length of sixth ray 103-134, of sixth from last ray 
82-98. Caudal fin 14-16, its length 165-242; its posterior margin changes 
shape considerably with size, being deeply forked in very small individuals and 
becoming emarginate or even slightly rounded in larger specimens. 

Pectoral fins 16-18, their length 222-275, extending posteriorly to above 
bases of rays 1-8 of anal fin; width of their bases 81-88. In larger specimens 
(100 mm. or more) the upper rays are longest and cause the posterior margin 
to be obliquely truncate or slightly falciform; the lower posterior margin is 
rounded. Pelvic fins rather short, their length 166-216, third rays longest, not 
reaching posteriorly to base of anal fin; inserted 232-312 from origin of anal 
fin, not entirely to entirely in advance of bases of pectoral fins. 

Upper lateral line 36-46, separated from origin of second dorsal fin by 6-10 
scale rows, ending below rays 3-6 from last of second dorsal fin; middle lateral 
line 5-14. The pores of the cephalic canals are small and often difficult to 
see, but are otherwise normal. Preoperculo-mandibular canals with 10-11 pores; 
infraorbital canals with 8 9 pores; supraorbital canals each with 4 pores and 
sharing a median coronal pore; temporal canals with 6 pores; supratemporal 
canal with 3-4 pores. 

Most scales on body nonctenoid, 47-64 in a lateral longitudinal series, 23-28 
rows around caudal peduncle; ctenoid scales present in area of sides covered 
by appressed pectoral fins. These latter have a single row of weak teeth along 
the posterior margin which is vertical and straight and may be recessed into the 
scale. There may be also a few weak projections on other scales of the body. 
Scales extend onto base of caudal fin and exposed bases of pectoral fins. Medial 
bases of pectoral fins, including small portions of body posterior to bases, naked; 
a small scaleless area also present on exposed side just anterior to base of rays. 
Head nearly entirely naked; small patches of scales present behind eyes, on 
uppermost part of operculum, and at postero-lateral parts of top of head. Round 
fleshy papillae cover remainder of top of head, and are present around lower 
and posterior parts of eyes, on snout, opercles, and sometimes on skin covering 
posterior parts of maxillaries. 

The color patterns of preserved specimens seem to vary considerably. Most 
of the specimens examined show no striking patterns anywhere, being darker 
above (bluish-grey to warm brown) shading to paler ventrally. The vertical 
fins are dusky, with pigment on both rays and membrane in the dorsal and anal 
fins, but mainly on the rays in the caudal fin. The pectoral fins are more or less 
dusky, being darkest in the more recently caught specimens. However, the 2 



Vol. XXXVIII] DeWITT: NOTOTHENIA FROM NEW ZEALAND 307 

specimens from Marion Island in the collection of the Zoological Institute in 
Leningrad have a strikingly different coloration. Overall they are brown, darker 
above, lighter below, with small spots and mottlings, more or less distinct, on 
the upper parts of the body. There are very clear spots and vermiculations on 
the top and sides of the head, including, in the larger specimen, most of the 
snout and the upper medial part of the upper lip. Rather irregular spots and 
stripes are present on the dorsal and caudal fins, and there is faint spotting of 
the upper pectoral rays in the larger specimen. This spotted coloration is very 
similar to that found on most specimens of N. angustata. Norman (1937b) 
adds the following: "... more or less distinct longitudinal stripes or series of 
spots on the sides; traces of oblique stripes below eye; . . . soft dorsal dusky, 
sometimes reticulated, and with a narrow pale margin. The young are more 
silvery, especially on the lower parts of head and body, and the fins are much 
paler." Waite (1916) gives a good description of specimens from Macquarie 
Island: "The general color is olive grey, the lower parts yellow; the markings 
are black and somewhat irregular, but two oblique bands may be traced below 
the eye; a branch from the upper one crossing the lower part of the opercle; 
the rest of the upper parts and sides of the head bear irregular spots and lines; 
six or seven bands cross the back to below the lateral line, whence they break 
and form blotches alternating with the bands. The first dorsal is dark and 
clouded; the second has a dark intramarginal band and a white edge; diagonal 
bars cross the lower portion, and the clouding leaves lacunae in the membrane; 
the anal is sooty, but the tips of the rays are lighter; the other fins are also 
sooty but without markings." 

In life the colors appear to be striking, as several authors have noted them. 
The back may be dark brown, dark grey-green, blue-grey, or rich golden-brown, 
passing to golden-yellow, cream, or reddish on the belly (the 189 mm. specimen 
in the Zoological Institute, Leningrad, was orange ventrally in life). The 
branchiostegal membranes may be bright orange-red or orange-yellow. The 
underparts of the head may be white, or the throat and jaws may be bright 
orange-red. The dorsal fins are blue-grey, the other fins grey (Cunningham, 
1871; Lonnberg, 1907; Norman, 1937b; Studer, 1879). 

The existence of pelagic juveniles in this species, which have been collected 
some distance from land over great depths, explains satisfactorily the wide dis- 
tribution of the species and the apparent lack of differentiation between the 
many seemingly isolated populations. In their general coloration they resemble 
closely the pelagic young of N. corikeps and A", rossii, species which also have 
wide distributions. 

Antarctic specimens. Among the Notothenia material which I have ex- 
amined are 2 large specimens captured well within the Antarctic Zone (Norman, 
1938; Andriashev, 1965) which appear to belong with A', magellanica. One, 



308 CALIFORNIA ACADEMY OF SCIENCES IProc. 4th Ser. 

USNM 171000, is from the Ross Sea and the other, in the Zoological Institute 
in Leningrad, is from the Scotia Sea. Both were collected from near the surface 
over fairly deep water. These 2 specimens thus present a problem with respect 
to both habitat and distribution. Information from the literature indicates that, 
except for the pelagic juveniles. A", magellanica is a near shore bottom fish, living 
among kelp, and that it can be captured with traps, hand lines and seines. Its 
distribution is primarily Subantarctic, extending into the edge of the Antarctic 
Zone only at Kerguelen and Macquarie islands. Further, very few species are 
known to inhabit both the Subantarctic and Antarctic zones. For these reasons 
the pelagic habit and high Antarctic localities of these 2 specimens suggest that 
they represent a different species. However, for nearly every character ex- 
amined they show no differences from Subantarctic material of A^. magellanica, 
and it may be that the observed differences are products of their large size. 
Also, since A^. magellanica is known to penetrate into the edge of the Antarctic 
Zone, it may prove to be one more species which inhabits both the Subantarctic 
and high Antarctic for at least part of the year. For the present, then, I shall 
consider these specimens as possible representatives of a differing population 
of A", magellanica of unknown taxonomic rank. 

Table 1 presents the pertinent measurements and counts taken from the 
Antarctic specimens together with the ranges of the measurements expressed as 
thousandths of the Standard Length. Comparison of these data with those taken 
from the subantarctic material shows that the Antarctic specimens have smaller 
eyes, a shorter distance between the tip of the snout and the nostrils, a greater 
distance between the nostrils and the edge of the orbit, a wider interorbital space, 
a shorter upper jaw, a deeper body, a greater distance between the origins of 
the second dorsal and anal fins, shorter pectoral and pelvic fins, and more rows 
of scales about the caudal peduncle. 

Besides the above differences, the lowermost gill rakers in the anterior series 
of the first gill arch are dentigerous and appear similar to those of the posterior 
series; the caudal fin is distinctly emarginate and each lobe is pointed. Most 
striking, however, are the presence of ctenoid scales over most of the body. Those 
covered by, and just above and below, the appressed pectoral fins are strongly 
ctenoid, while those posteriorly on the sides of the body, anteriorly along the 
back, especially anterior to the first dorsal fin, and anterior to bases of pectoral 
fins are more weakly ctenoid. All of the scales on the belly, even anterior to the 
pelvic fins, are ctenoid. 

There are no obvious markings on the body or head. Top and sides of head 
and upper parts of body a dark grey -brown or bluish black; body shading to 
paler below, head becoming paler more abruptly along ventral edges of cheeks 
and opercles, and on lower jaw; the Scotia Sea specimen is very pale orange- 
pinkish below. First dorsal fin uniformly black; membranes of second dorsal 



Vol. XXXVIII] DeWITT: XOTOTHENIA FROM NEW ZEALAND 309 

Table 1. Measurements {in mm.) and counts from two Antarctic specimens of Xotothenia 
magellanica, with ranges of measurements expressed as thousandths of the Standard Length. 





USNM 


ZIL 




Observation 


171000 


Specimen 


Range 


Measurements 








Standard Length (SL) 


229 


261 




Length of head (HL) 


64.3 


68.8 


253-281 


Width of head (HW) 


51.6 


— 


215 


Orbital diameter (0) 


12.2 


10.1 


39-53 


Length of snout (Sn) 


19.8 


22.1 


85-86 


Snout to nostril distance (Sn-N) 


11.1 


lO.S 


40-48 


Nostril to nostril distance (N-N) 


16.3 


18.4 


71 


Interorbital width (10) 


32.5 


3S.6 


136-142 


Length of postorbital part of head (PO) 


37.1 


— 


162 


Length of upper jaw (JL) 


21.2 


24.0 


92-93 


Length of caudal peduncle (CPL) 


27.7 


34.9 


121-134 


Depth of caudal peduncle (CPD) 


22.1 


21.9 


84-96 


Dorsal to caudal distance (D-C) 


28.6 


— 


125 


Depth of body (BD) 


72.6 


70.1 


269-317 


Pectoral to pectoral distance (P-P) 


50.2 


— 


219 


Second dorsal to anal distance (Dj-A) 


86.7 


— 


379 


Snout to first dorsal (Sn-Di) 


70.5 


70.8 


271-308 


Snout to second dorsal (Sn-D-) 


91.7 


99.9 


383-400 


Snout to anal (Sn-A) 


126 


141 


540-550 


Anal to pelvic distance (A-V) 


71.5 


83.5 


312-320 


Length of caudal fin (CL) 


46.7 


48.S 


186-204 


Length of pectoral fin (PL) 


49.9 


57.0 


218 


Length of pelvic fin (VL) 


35.8 


39.8 


152-156 


Length of sixth ray of second dorsal fin 


30.4 


— 


133 


Length of sixth ray of anal fin 


26.4 


— 


115 


Counts 








First dorsal fin (Di) 


4 


4 




Second dorsal fin (D2) 


31 


29 




Anal fin (A) 


25 


25 




Caudal fin (C) 


16 


16 




Pectoral fin (P) 


16 


16 




Lateral scales (LatSc) 


52 


S3 




Scale rows between lateral line and origin 








of second dorsal fin (LL-D:^) 


10 


— 




Scales around caudal peduncle (ScArCP) 


30 


32 




Upper lateral line (ULL) 


37 


44 & 45 




Middle lateral line (MLL) 


10 


2 




Branchiostegal rays (Br) 


6 


— 




Preoperculo-mandibular pores 


10 


12&10 




Infraorbital pores 


8 


9&8 




Supraorbital pores 


4 


4 




Temporal pores 


6 


6 




Supratemporal pores 


3 


3 




Anterior gill rakers, first arch 4+1 + 10=15 


4+1+11 = 16 





310 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

fin black, the rays pale hyaline, creating the effect of white stripes on a black 
field. Membranes of anal fin dusky basally, clear hyaline toward margin, the 
rays pale. Caudal fin dusky, especially along upper and lower edges of rays. 
Pectoral fins dusky along upper and posterior margins, paler centrally and 
below; pelvics dusky. 

Distribution. Notothenia magellanica has been recorded from the Magel- 
lanic region; Kerguelan, Macquarie, Aukland, and Campbell islands; and from 
the South Island of New Zealand. In addition it is recorded here for the first 
time from Marion Island and 2 localities well within the Antarctic Zone. Except 
for pelagic juveniles and the 2 far southern records, the species appears to in- 
habit only very shallow water, as all records (where the information is given) 
state that specimens were secured by traps, hand lines, or seines. The Discovery 
obtained a few juveniles with dip nets and tow nets from or near the surface in 
open waters. Kenny and Haysom (1962 ) state that the species lives among kelp 
near the shore at Macquarie Island, and Forster (in Bloch and Schneider, 1801; 
1844) states that about Tierra del Fuego it lives near the shore among sea weed. 

In the Magellanic region N . magellanica appears to be restricted to the west 
coasts of Tierra del Fuego and Patagonia, and the Falkland Islands, a pattern 
similar to that of several other Subantarctic species. It is probable that adults 
everywhere are associated with rocky and protected areas near shore. 

Discussion. Although Norman (1937b) listed Gadus magellanicus Forster 
(in Bloch and Schneider, 1801) with a sign of interrogation under Notothenia 
macrocephala Giinther, he considered Forster's description to be equally ap- 
plicable to A^. macrocephala and N . microlepidota (non N. microlepidota of 
Hutton, but equals A^. angustata of Hutton; see discussions under both species). 
His reasons for this position were that the unpublished drawing of the species 
by Forster is a rough sketch which, while definitely representing a Notothenia, 
is not of sufficient detail to identify the species, and that the anal fin is described 
as having 25 rays, a number common to both species. Through the courtesy of 
Mr. A. C. Wheeler of the British Museum (Natural History) and the Trustees 
of the British Museum I have been able to obtain a photograph of Forster's 
drawing which is reproduced here (fig. 1). Although the drawing is obviously 
unfinished, it shows definitely that TV. macrocephala is a synonym of Gadus 
magellanicus. The pectoral fin is drawn with an oblique posterior margin and 
with the upper rays longest, the snout is separated from the top of the head by 
an abrupt rounding above the nostrils, and the caudal fin is emarginate. These 
characters are diagnostic for the present species. 

Norman (1937b) also lists Notothenia porteri Delfin as a synonym of A^. 
magellanica, but a careful reading of the description demonstrates that the name 
is a synonym of A", angustata. A full discussion is presented under that species. 



Vol. XXXVIII] DeWITT: NOTOTHENIA FROM NEW ZEALAND 311 






0^ 

Figure 1. Notothenia magellanica. Reproduction of J. G. A. Forster's unpublished 
drawing of Gadus magellaniais, by permission of the Trustees of the British Museum 
(Natural History). 

A final problem has been the location and designation of the types of N. 
maoriensis, N . argnta, N . hasslcriana, and -V. antarctica. 

It would appear that the type of N . maoriensis has been lost. In 1965 Miss 
M. Bijchler (now Mrs. M. Darby), then Assistant Zoologist in the Canterbury 
Museum at Christchurch, New Zealand, informed me that an old register, dating 
back to the early part of this century or even into the last century, contains the 
following entry: 

''Notothenia maoriensis Haast, Trans. N. Z. Inst. vol. S, p. 276 N. coriiceps Hutton, Cat. 
Fishes N. Z.: 32 (nee Richardson) Stuffed (Type lost, originally stuffed)." 

Miss Biichler made a thorough search through the fish collection and catalogues 
and the above entry was the only positive result. Therefore it seems fairly 
certain that the type is no longer in existence. 

A number of fishes, some of which are types, were presented to the British 
Museum in the 1880's by the Otago Museum in Dunedin, New Zealand. Among 
them is a specimen labelled as the type of Notothenia argida. Its total length is 
about 179 mm., which is close to the length of 7% inches (equals 184 mm.) 
given by Hutton in the original description. Dr. D. R. Simmon of the Otago 
Museum informed me (letter dated 8 April 1964) that "although N. arguta is 
entered as a name in the register . . . there is no record of a specimen being held 
by this museum." I therefore conclude that the British Museum specimen is 
indeed the type. 

Notothenia hassleriana was described from an unknown number of specimens 
collected at 2 localities in the Strait of Magellan. I have examined 2 specimens 



312 



CALIFORNIA ACADEMY OF SCIENCES 



[Proc. 4tii Ser. 




' -' ^^ , ■:^/■^p^i^i;^;^^y^,^,^.^'iA 




Figure 2. Notothenia magellanica, from Steindachner, 1876. 



labeled as types, one each from the 2 localities. It is possible that these 2 
specimens are all that Steindachner had, for Dr. P. Kahsbauer of the Naturhis- 
torisches Museum in Vienna indicated in letters dated 7 October and 7 Novem- 
ber, 1964, that these were all he could find. In any event, the specimen from 
Port Gallant (register number 59926) is very similar to the illustration pub- 
lished by Steindachner and reproduced here (fig. 2), and I designate this 
specimen as the lectotype. 

Notothenia antarctica was described from a single specimen, 35 cm. long, 
collected by Dr. Studer during the voyage of the SMS Gazelle from Accessible 
Bay, Kerguelen Island. Dr. C. Karrer of the Zoologisches Museum of Humboldt 
University in Berlin has written that a specimen identified as N . antarctica of 
the proper size, from the above locality and collected by the Gazelle is in the 
fish collection there. Although it is not labeled as the type, it is undoubtedly 
the specimen Peters used for his description. Professor Kurt Deckert of the same 
museum had earlier written that although the register of the fish collection listed 
the type of N . antarctica, he had been unable to find it. 

Notothenia rossii Richardson. 

Notothenia rossii Richardson, 1844: 9-10, pi. 5, figs. 1 & 2 (original description and illustra- 
tion; type locality unknown, but probably the Kerguelen Islands (Regan, 1916); type 
lost) ; Gunther, 1860: 263 (description) ; Norman, 1937a: 61, 64 (description, separa- 
tion from N. coriiceps) ; Norman, 1938: 25 (description, illustration, distribution) ; 
Blanc, 1951: 495 (listed, food); Blanc, 1954: 191 (listed); Blanc, 1958: 137 (listed, 
illustration); Blanc, 1961: 123-124 (description); Bellisio, 1966: 69, foto 40 (listed, 
illustration). 

Notothenia rossi Regan, 1913: 240, 276-277 (description); Andriashev and Tokarev, 1958: 
199 (juvenile listed). 

Macronotothen rossii Gill, 1862: 521 (listed). 

Notothenia marmorata Fischer, 1885: 53-55 (original description; type locality South 
Georgia, probably at about 54°31'S., 36°05'W. ; types (2 specimens remain of original 



Vol. XXXVIII] DeWITT: NOTOTHENIA FROM NEW ZEALAND 



313 



Table 2. Measurements (in mm.) and counts from the types of Notothenia macrocephala, 
N. arguta and N. hassleriana. Abbreviations are as in table 1, with the addition of body 
width (BW), anterior gill rakers of first gill arch (AntGR) and longest pelvic rays (LongVR). 
Where two measurements or counts are given, the second is taken from the right side. 



Observation 


\. macrocephala: 

BMXH 1860.2.20.2 

(Type) 


iV. hassleriana: 
NMV 59926 
(Lectotype) 


;V. hassleriana: 

NMV 65389 
( I'aralectotype) 


N. arguta: 

BMNH 1886.11.18.28 

(Type) 


SL 


__ 


86.0 


87.0 


ISO 


HL 


84 


24.S 


25.4 


45.4 





16 


5.7 


6.1 


9.5 


Sn 


— 


7.5 


7.1 


13.4 


Sn-N 


— 


4.9 


4.5 


8.7 


N-N 


— 


4.9 


5.2 


9.9 


10 


36 


8.6 


9.0 


17.6 


PO 


— 


13.1 


13.9 


23.9 


JL 


, — 


9.0 


8.6 


16.4 


CPL 


32 


10.5 


10.3 


16.7 


CPD 


— 


8.3 


8.4 


14.7 


BD 


— 


21.6 


22.3 


38.7 


BW 


— 


10.5 


12.3 


22.1 


P-P 


— 


14.8 


17.6 


30.9 


Sn-Di 


— 


27.5 


27.4 


49.0 


Sn-D, 


— 


36.7 


36.3 


63.8 


Sn-A 


— 


45.8 


45.8 


77.5 


A-V 


— 


26.7 


24.6 


40.5 


CL 


— 


18.4 


18.0 


29.4 


PL 


— 


21.0 


20.1 


38.1 


VL 


— 


16.8 


16.8 


29.8 


AntGR 


— 


5 + 1 + 11 = 17 


5 + 1 + 10=16 


6 + 1 + 11 = 18 


Di 


— 


4 


5 


4 


D, 


31 


30 


31 


30 


A 


24 


24 


24 


24 


C 


— 


16 


16 


16 


P 


17 


16& 17 


16&16 


17 


LongVR 


— 


3 


3 


3 


LatSc 


49 


SI 


54 


SO 


LL-D, 


— 


7 


6 


7 


ScArCP 


— 


27 


24 


24 


ULL 


38 


41&42 


40 & 38 


38&36 


MLL 


11 


9&11 


lO&ll 


9 (?) &10 



3) in Hamburgischen Zoologischen Staatsinstituts und Zoologischen Museums, Ham- 
burg) . 

Notothenia macrocephala marmorata Lonnberg, 1905: 34-36, 53 (description, spawning); 
LoNNBERG, 1906: 94-95 (description, spawning, food). 

Notothenia coriiceps var. macquariensis Waite, 1916: 64-66, pi. 5, fig. 3 (original description 
and illustration; type locality Macquarie Island; lectotype in South Australian Museum, 
Adelaide); Regan, 1916: 378 (differentiation from N. coriiceps); Norman, 1937a: 60-61 
(synonymized) . 



314 CALIFORNIA ACADEMY OF SCIENCES IProc. 4th Ser. 




Figure 3. Notothenia rossii, from Waite, 1916. 

Notothenia rossii marmorata Nybelin, 1947: 22-26 (differentiation from A'^. r. rossii, de- 
scription, distribution); Nybelin, 1951: 23-27 (description, differentiation from N. r. 
rossii, spawning) ; Olsen, 1954: 373-382 (description, growth, food, habits) ; Ruud, 
1954: 849 (oxygen capacity of blood) ; Olsen, 1955: 88 (biology compared to Chan- 
nichthyids) ; Ladiges, Wahlert, and Mohr, 1958: 165 (designation of lectotype) ; 
Andriashev, 1959: 5 (vertebrae). 

Notothenia rossii rossii Andriashev, 1959: 5 (vertebrae). 

Material examined. SU 67031: washed up on beach, Macquarie Island (1; 

461 mm.; partly eaten and eviscerated). 
SAM (uncatalogued) : Macquarie Island (1 ; 342 mm.; lectotype of A^. coriiceps 

macquariensis) . 

The following material has been examined for purposes of comparison with 
the above Macquarie Island specimens. 

BMNH 1937.7.12.563 4: Jetty (probably Government Jetty, Grytviken), 

South Georgia (2; 129 & 143 mm.). 
USNM 107158: Stromness Harbour, South Georgia (1; 208 mm.). 
USNM 179080: King George Island, South Shetland Islands (1; 274 mm.). 
VSC-Eltanin Station 671: South-west of South Georgia Island, 54°41''S., 

38°38'W.; 220-320 m.; 10-foot Blake trawl (1; 432 mm.). 

Description 
Body less deep than N. magellanka, not becoming much deeper than head; 
ventral profile curves more evenly than dorsal profile, which rises most steeply 
in the snout; body compressed posteriorly, but becomes somewhat depressed 
anteriorly. Length of head 319-323, its width 234-292, its depth 204; depth 
of body 213, its width 138, dorsal to anal distance 235; pectoral to pectoral 
distance 219-228; length of caudal peduncle 103-104, its depth 85-91; dorsal 
to caudal distance 105-109. Vertebrae 20 + 31 = 51. 



Vol. XXXVIII] DeWITT: NOTOTHENIA FROM NEW ZEALAND 315 

Snout rises steeply in a smooth curve from lateral view, its length 86-95. 
Nostrils short tubes with the posterior margin raised into a point, placed 53-63 
from tip of snout, 24-25 from orbits, and 63-67 apart. The mouth appears 
larger than in N. magellanica, but the upper jaw extends posteriorly only under 
anterior edge of pupil; length of upper jaw 130-138; lower jaw projects slightly 
beyond upper jaw. Eyes directed laterally, placed just below dorsal profile of 
head; diameter of orbit 51-58. Interorbital region broad and almost flat, both 
from lateral and frontal views, its width 102-105. Length of postorbital part 
of head 174-186. 

Teeth in both jaws in 2 bands; outer band a single row of somewhat enlarged, 
evenly spaced, canine-like teeth extending almost full length of jaw, becoming 
slightly smaller anteriorly and absent near symphysis; inner band lies immedi- 
ately behind outer row, broad anteriorly, becoming narrow posteriorly, extending 
posteriorly only % to Vj length of jaw (upper jaw) or as far as outer row (lower 
jaw). Tongue free anteriorly, fleshy, but not soft. Oral membranes extend most 
of length of jaws, papillose only along anterior edges. 

Larger gill rakers in anterior series of first gill arch flattened, nondentigerous, 
and not very elongate, arranged 6 + 0-1 -|- 13-14 = 20. Posterior gill rakers 
of anterior gill arch dentigerous distally on anterior face, arranged 1 + 1 + 11 = 
13 (SAM specimen); gill rakers of remaining arches similar. Branchiostegal 
rays 6; pseudobranchiae curved ventralward posteriorly. 

First dorsal fin 5-6, lower than second dorsal fin, length of longest spine 
39-60; its origin 301-332 from tip of snout, from slightly behind to slightly in 
advance of upper end of base of pectoral fin. Second dorsal fin 32-33 (Waite, 
1916, gives 33-34), its origin 406-451 from tip of snout, 31-51 from base of 
last ray of first dorsal fin; first ray short, heavy and unbranched but segmented; 
length of sixth ray 101-102, of sixth from last ray 70-86. Anal fin 27-28, its 
origin 558-562 from tip of snout, beneath bases of rays 8-10 of second dorsal 
fin; length of sixth ray 81-92, of sixth from last ray 69-77. Caudal fin 14-16, 
its length 154-173, its posterior margin truncate. 

Pectoral fins 22-23, their length 215-223, reaching posteriorly to above base 
of first ray of anal fin or not reaching to anal fin; middle portion of posterior 
edge truncate, upper and lower portions rounded; uppermost ray very short, 
about 48. Pelvic fins inserted 237-330 from origin of anal fin, entirely in ad- 
vance of bases of pectoral fins; their length 157-160, third, or third and fourth 
rays longest, not at all reaching to anal fin. 

Upper lateral line of body with 40-57 tubular scales, dipping slightly above 
upper end of base of pectoral fin, ending posteriorly below about fifth to ninth 
from last rays of second dorsal fin, and separated by about 6-7 scale rows from 
origin of second dorsal fin. Middle lateral line with 15-17 tubular scales, 
originating below or a little behind end of upper lateral line and extending a 



316 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

short distance onto base of caudal fin. Cephalic lateral line system normal in 
pattern, but pores very small and difficult to see. Preoperculo-mandibular canals 
with 10 pores, not connected to temporal canals; infraorbital canals with 8-9 
pores; supraorbital canals with 4 pores and sharing a coronal pore; temporal 
canals with 6 pores; supratemporal canal with 3 pores. 

Scales in lateral longitudinal series 55-57; 28-29 around caudal peduncle. 
Scales nonctenoid except for those in area of side of body covered by appressed 
pectoral fin and a little posteriorly which are weakly ctenoid; scales present 
everywhere on body except medial (posterior) base of pectoral fin and area im- 
mediately adjacent, and an arc along lateral base of pectoral fin; scales extend 
onto proximal part of caudal fin and onto lateral proximal part of pectoral fin; 
scales small on belly, ventral area anterior to pelvic fins and on back anterior to 
first dorsal fin. Scales absent on head except for about upper VL' of cheeks behind 
eyes, about upper V-i of operculum, and 2 small patches on each side, one in front 
of the other, on posterolateral corners of top of head. Head only very slightly 
rugose, with small raised vermiculations, the most prominent radiating from 
eyes. Low ridges present, probably associated with parietal and pterotic bones. 

Color (in alcohol) of the SAM specimen is dark grey-brown with some blue 
above, becoming lighter, somewhat yellowish below. Second dorsal fin with dark 
longitudinal bands, 3 anteriorly, 2 posteriorly, rather irregular anteriorly. The 
SU specimen is brownish black above, lighter on belly. The second dorsal is 
marked with somewhat irregular brownish bands which extend posteroventrally 
in anterior part of fin and more or less parallel with back in posterior part of fin. 
Anal fin dusky except for a pale margin; caudal fin irregularly and indistinctly 
mottled. Two faint stripes on head, one extending along edge of upper jaw, the 
other extending from posteroventral edge of eye to angle of preopercular. 

Subspecies. Notothcnia marmorata Fischer, described from South Georgia, 
has long been considered a synonym of .V. rossii since comparison of material 
from Kerguelen and Macquarie islands with that from the region of the Scotia 
Sea has shown that the two populations are very similar. Nybelin (1947, 1951) 
was the first to call attention to differences between specimens from the two 
regions, and he resurrected the name ^' marmorata'^ as a subspecies of N . rossii. 
Unfortunately, very little material has been reported from the Kerguelen- 
Macquarie region (Richardson, 1844; Waite, 1916; Blanc, 1951, 1954, 1961). 
The most reliable published information is that by Richardson; Waite's 1916 
paper contains numerous errors, and his methods of counting differ in some 
instances from those now in practice; the data given in the papers by Blanc 
seem to have been copied from reports on Antarctic material and cannot be 
used. For these reasons the data presented above, although obtained from only 2 
specimens, are important additions to our knowledge of the species. 

Combining my observations with those of Richardson, it seems possible that 



Vol. XXXVIII] DeWITT: XOTOTHEXIA FROM NEW ZEALAND 



317 



Table 3. Measurements {in mm.) and counts from two specimens of Xotothenia rossii 
from Macquarie Island. Abbreviations are as in table 2. Where two measurements or counts 
are given, the second is taken from the right side. 



Observation 


Lectotype 


SU 6 7031 


Observation 


Lectotype 


SL" 67031 


SL 


342 


461 


A-V 


98.1 


152 


HL 


109 


149 


CL 


S2.7 


79.8 





17.5 


26.8 


PL 


73.5 


103 


Sn 


32.5 


39.7 


VL 


54.9 


72.3 


Sn-N 


21.6 


24.4 


AntGR 


6 + 1 + 1.3 


6+0+14 


X-N 


22.8 


29.1 




= 20 


=20 


10 


36.0 


47.1 


Di 


6 


5 


PO 


59.5 


85.7 


Do 


33 


32 


JL 


44.4 


63.4 


A 


28 


27 


CPL 


35.1 


48.1 


C 


16 


14 


CPD 


31.0 


39.3 


P 


22&22 


23 


BD 


72.3 


— 


LongVR 


3 


3 and 3 & 4 


BW 


47.1 


— 


LatSc 


57 


55 


P-P 


74.9 


105 


LL-D, 


7 


6 


Sn-Di 


103 


153 


ScArCP 


28 


29 


Sn-D. 


139 


208 


ULL 


57&54 


40 


Sn-A 


191 


259 


MLL 


17&17 


15 



the Kerguelen and Macquarie material differ from the Scotia Sea material in 
having a longer snout (86-95 vs. 75-86), a broader interorbit (102-105 vs. 
87-100), a longer upper jaw (130-138 vs. 106-123), a greater distance between 
the tip of the snout and the origin of the anal fin (558-562 vs. 472-547), fewer 
rays in the second dorsal fin {2>2-2)2i vs. 34-35), fewer vertebrae (51 vs. 52-53), 
and different coloration. It may be that the proportional differences are due 
to size, as the 2 Macquarie Island specimens examined are larger than nearly all 
of those seen from the Scotian region, but the counts and color differences seem 
to be reliable. In color, specimens from the Scotian region have the sides of the 
body covered by a series of irregular lines and blotches, with sometimes a dark 
arc at the base of each pectoral fin and spots on top of the head. The second 
dorsal is marked in much the same manner as in the Macquarie specimens, but 
the bands are much more distinct. 

For the above reasons I believe there is good evidence for following Nybelin 
in recognizing as subspecies two populations, one, N . rossii rossii, inhabiting the 
Kerguelen and Macquarie islands, and the other, X . rossii marmorata. inhabiting 
the islands of the Scotia Ridge system, including the South Shetland Islands. 

Discussion. In his original description of N . coriiceps var. tnacquariensis, 
Waite stated that the type was in the South Australian Museum, but he did not 
specifically designate either of the 2 specimens upon which his description was 
based. I therefore select the specimen from the South Australian Museum listed 



318 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

under material examined as the lectotype. The second specimen, now presumably 
in the Australian Museum, Sydney, becomes a paralectotype. I do not know 
whether the lectotype is the specimen illustrated by Waite, but selecting the 
specimen in the South Australian Museum accords with Recommendation 74D 
of the International Code which suggests that a lectotype be selected from the 
material in the institution containing the largest number of types from the 
collection worked upon by the original author. 

Nototheiiia angustata Hutton. 

Notothenia coriiceps (non Richardson) Hutton, 1872: 26 (brief description); Thompson 
and Anderton, 1921: 94 (listed). 

Notothenia cornucola (non Richardson) Hutton, 1873: 262-263 (brief description). 

Notothenia angustata Hutton, 1875: 315-316 (original description; type locality Dunedin 
Harbour; type in Otago Museum); Hutton, 1876: 213 (an almost verbatim reprint of 
the previous paper; localities given as Dunedin and Bluff harbours) ; Hutton, 1879: 339 
(listed, synonymy); Hutton, 1890: 279 (listed); Gill, 1893: 118 (listed); Waite, 1907: 
30 (listed). 

Notothenia parva Hutton, 1879: 339 (original description; type locality Auckland Islands; 
types in Dominion, Otago and British Museums) ; Hutton, 1890: 280 (listed) ; Gill, 
1893: 118 (listed); Waite, 1907: 30 (listed). 

Notothenia porteri Delfin, 1899b: 118-120 (original description; type locality Talcahuano, 
Chile; type (or types) possibly in the old natural history museum in Valparaiso, Chile). 

Notothenia microlepidota (non Hutton) Boulenger, 1902: 185 (listed); Waite, 1909: 590- 
594 (description, illustration); Regan, 1913: 277-278 (description); Waite, 1916: 69 
(listed); Regan, 1916: 379 (synonymy); Thompson and Anderton, 1921: 94 (listed); 
MacDonagh, 1936: 428-429 (synonymy) ; Norman, 1937b: 90-91 (description, syn- 
onymy, distribution); Fowler, 1951: 314 (key); Moreland, 1957: 34 (listed); Parrott, 
1958: 110-111 (description, variation). 

Notothenia latifrons Thompson, 1916: 434-435, pi. 3, fig. 1 (original description and illustra- 
tion; type locahty Sandy Point (Punta Arenas), Strait of Magellan; holotype in U.S. 
National Museum). 

Notothenia macrocephala (non Giinther) Fowler, 1926: 283 (description). 

Notothenia patagonica MacDonagh, 1931: 100 (original description; type locahty among 
the rock ledges of Bahia del Fondo, Golfo San Jorge, Santa Cruz (province), Patagonia 
(Argentina) ; holotype in Museo de La Plata) ; MacDonagh, 1934: 84-91, pi. 10, figs. 
2 & 3, pi. 11, figs. 1 & 2, pi. 12 (description, illustrations, scales, systematics) . 

Material Examined. BMNH 1886.11.18.29: Auckland Islands; from the 
Otago Museum (1; 71.7 mm.; lectotype of N. parva). 

BMNH 1886.11.18.30: Dunedin, New Zealand (1 ; 238 mm.). 

BMNH 1936.7.7.4: among the rocky ledges of Bahia del Fondo, Golfo San 
Jorge, Santa Cruz (province), Patagonia (Argentina) (1; 230 mm.; paratype 
of N . patagonica) . 

USNM 39670: New Zealand ( 1 ; 182 mm.). 

USNM 176391: Huiches Island, Chile (45°10'30"S., 73°33'W.) (1; 332 mm.). 

MLP 12-XII-30-1: same data as for BMNH 1936.7.7.4 (1 ; 315 mm.; Holo- 
type of N . patagonica) . 



Vol. XXXVIII] DeWITT: XOTOTHEXIA FROM NEW ZEALAND 



319 




Figure 4. Notothenia angiistata, from Waite, 1909. 



CM (uncatalogued): Ranui Cove, Auckland Island (2; 56.2 and 85.8 mm.). 

Selected measurements and counts were taken from the following specimens, 
all deposited in the Dominion ^Museum, New Zealand. 2864: Campbell Island 
(1); 2897: Oreti Beach, Southland (1); 3124: Campbell Island (1); 3332: 
outer Ranui Cove, Auckland Island (7); uncatalogued: Waitangi, Chatham 
Island, 43°36.2'S., 176^48. 5'\V. (3); uncatalogued: Glory Bay, Pit Island, 
Chatham Islands, shore, 43°47'S., 179°30'W. (1). 

Material in the Canterbury Museum (all uncatalogued) from the following 
localities was also examined. Tucker Point, Port Ross, Auckland Island, under 
stones (2); west coast of Campbell Island (1); Tucker Cove, Campbell Island, 
among kelp at low tide (1); Auckland Islands (2); Laurie Harbour, Auckland 
Island ( 1 ) ; Ranui Cove, Auckland Island ( 1 ) . 

Description. Larger specimens more than usually compressed posteriorly; 
caudal peduncle distinctly deeper than long. Smaller specimens more cylindrical; 
caudal peduncle may be longer than deep. In region of bases of pectoral and 
pelvic fins, body becomes broader and less deep; the head appears depressed 
and small, although its measured length is similar to those for other species. 
Dorsal and ventral profiles about equally convex, or the ventral profile, at 
least of head, may be a little more convex than dorsal profile. Length of head 
302-345, its width 187-281, its depth 181-187; depth of body 197-259, its 
width 140-188; dorsal to anal distance 229-288, pectoral to pectoral distance 
189-266; length of caudal peduncle 92-117, its depth 92 125; dorsal to caudal 
distance 89-125. Vertebrae 17-18 + 27-29 = 44-46. 

Snout broad and flattened, its length 84-98, longer than diameter of orbit. 
In lateral view the snout appears short, but its breadth causes its measured length 
to be larger. A pair of ridges, through which the supraorbital canals extend, 
separate the medial and lateral parts of the snout. These ridges curve nround 



320 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

the posterior and medial sides of the nostrils and extend anteriorly to end at 
the edge of the groove behind the upper lip. In larger specimens the medial 
portion of the snout is flat and somewhat raised ; in smaller specimens the ridges 
are little developed and the snout is more evenly rounded. The nasal tubes lie 
in shallow depressions, placed 52-66 from tip of snout, 17-30 from orbit, and 
53-64 apart. The posterior half of each tube is raised into a pointed flap which 
can be used to constrict or close the nasal opening. 

Eyes rather small, diameter of orbit 45-82, placed entirely within upper 
half of side of head, either above or extending slightly below line between tip 
of snout and upper end of base of pectoral fin, not projecting into dorsal profile 
of head. Interorbital region very broad, its width 81-104. Ridges of supra- 
orbital canals, described above for snout, are continued through interorbital 
region on each side above eyes; these ridges are clearly visible on the small 
specimens. Medial portion of interorbital space flat and covered with elongate 
or finger-like papillae. 

Supraorbital ridges continue onto postorbital part of head as ridges of 
temporal canals, and extend to above operculum. Upper surface of head behind 
eyes almost flat, covered with papillae like those of interorbital region; posterior 
limit of papillae follows posterior line of head medially, but overlies post- 
temporal bone laterally. Length of postorbital part of head 167-198. 

Mouth broad, somewhat oblique, lower jaw projecting slightly; length of 
upper jaw 123-150, maxillary extending under anterior Vi to % of eye; width 
of jaws 176-180. Teeth all conical, arranged in 2 bands in each jaw. Outer 
bands uniserial, composed of enlarged, almost canine-like teeth; inner bands 
broader, especially anteriorly, composed of smaller and more slender teeth. 
Inner band of lower jaw extends only along anterior V2 of jaw; that of upper jaw 
extends about full length of jaw. Outer bands of both jaws extend about full 
length of jaws, that of upper jaw being slightly longer than inner band. 

Gill rakers in anterior series of first gill arch short, blunt, somewhat flattened 
obliquely to long axis of arch and bearing teeth distally; arranged 5-7 + 0-1 + 
11-15 = 17-22; in smaller specimens those near angle may be more elongate, 
bearing teeth along upper edges. Gill rakers of posterior series of first gill arch 
dentigerous and only slightly flattened at right angles to long axis of gill arch, 
arranged 0-1 + 0-1 + 10-11 = 11-13; gill rakers of remaining arches similar. 
Branchiostegal rays 6. 

First dorsal fin 4-7, its origin 289-329 from tip of snout, above or slightly 
in advance of upper end of base of pectoral fin; second spine longest, its length 
62-102. Second dorsal fin 27-30, its origin 400 457 from tip of snout and 
17-75 from base of last spine of first dorsal fin; length of sixth ray 111-162, 
of sixth from last ray 97-122. Anal fin 22-26, its origin 505-577 from tip of 
snout, originating below bases of sixth to eighth rays of second dorsal fin ; length 



Vol. XXXVIIl] DeWITT: NOTOTHENIA FROM NEW ZEALAND 321 

of sixth ray 96-138, of sixth from last ray 85-105. Caudal fin 14-16, its length 
157-219, its posterior margin very slightly rounded, almost truncate. 

Pectoral fins 17-19, their length 193-240, not extending to origin of or 
reaching to above first four rays of anal fin, their posterior margins rounded; 
width of their bases 80-97. Pelvic fins placed 262-346 from origin of anal fin, 
entirely in advance of bases of pectoral fins, their length 174-217, third or 
fourth rays longest, not reaching posteriorly to origin of anal fin. 

Upper lateral line 45-61, terminating from below fourth from last ray to 
slightly behind posterior end of base of second dorsal fin, separated from origin 
of latter by 6-7 scale rows. Middle lateral line 9-18, extending a short distance 
onto base of caudal fin. Cephalic lateral line canals of normal pattern, the pores 
very small and difficult to see. Preoperculo-mandibular canals with 9-10 pores, 
connected to temporal canals; infraorbital canals with 8-10 pores; supraorbital 
canals each with 4 pores and sharing a median coronal pore ; temporal canals with 
5-6 pores; supratemporal canal normally with 3 pores, but in 1 specimen the 
canal is incomplete across the head and consists of a short tube on each side ex- 
tending dorso-medially from the temporal canals, each with a single pore at 
its end. 

Most scales on body ctenoid, 49-60 in a lateral longitudinal series, 27-31 
rows around the caudal peduncle. Parrott (1958) records 61-69 scales in a 
lateral longitudinal series, but none of the specimens I have examined had counts 
that high; MacDonagh (1931) gives a lateral scale count of 68 for the holotype 
of N. patagonica, but I count only 60; Thompson (1916) records 67-73 lateral 
scales. Since Thompson's counts were made along the lateral line, and above it, 
from the angle of the operculum to the base of the caudal fin, higher counts 
would be expected. It is probable that the other high counts were made in a 
similar manner. Xonctenoid scales present on belly and area anterior to bases of 
pectoral fins; a few may be found along bases of dorsal and anal fins. Scales 
extend onto basal parts of caudal fin and, except for a narrow naked crescent at 
bases of pectoral rays, onto lateral bases of pectoral fins. Scales absent directly 
in front of bases of pelvic fins, but medially they extend to area covered by 
fold of branchiostegal membrane across isthmus. 

Head almost entirely naked; a few scales, some of which may be ctenoid, 
present at posterolateral corners of head above temporal canals and on either 
side of supratemporal canal; a larger patch, some being ctenoid, present on 
upper part of operculum; a still larger patch, all nonctenoid, present on upper 
part of cheek behind eye; a few scattered scales may be present below eye. 

Ground color of BMNH specimen 1886.11.18.30 brown, somewhat lighter, 
perhaps originally white or yellowish, on belly. Head somewhat darker and 
greyish above. No prominent markings present on body. Sides of head with 
darker vermiculations creating a marbled appearance; these markings continued 



322 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

onto lateral parts of snout, lips, lower jaw, and faintly onto branchiostegal rays. 
All vertical fins more or less uniformly brownish-dusky; second dorsal fin with 
indistinct and irregular darker brown markings on rays; anal fin with 1 or 2 
series of darker markings on rays, tending to form horizontal lines. Rays of pec- 
toral fins with brown spots, arranged to form bars on left side, but irregular on 
right side; pelvic fins with faint marbling similar to that on sides of head. 

The USNM specimen from New Zealand is nearly entirely a uniform dark 
grey-brown. The larger of the 2 Canterbury Museum specimens has on the 
body irregular light areas over a dark background. The head is uniformly dark 
above and on the snout, but on each cheek is a series of 4 light lines, partially 
broken into spots, which radiate from the ventral and posterior parts of each 
eye; irregular light spots are present on the operculum. The vertical fins are 
generally dark; the second dorsal fin has 1 to 3 light spots along its rays creating 
horizontal lines, most distinct anteriorly and basally; the anal fin shows light 
areas, irregularly arranged anteriorly, horizontally arranged posteriorly; tips of 
rays of second dorsal and anal fins pale. The pectoral fins show only faint 
barring and spotting. The smaller specimen is essentially the same as the larger, 
but the light areas on the sides of the head are larger and less broken. 

Little has been recorded of life colors. Hutton (1875; p. 316) gives "Variable 
in color from dark olivaceous black to olive-green, slightly mottled with blackish 
on the back; lips speckled with white; axil of pectorals yellow; caudal and dorsal 
blackish." In his description of N. porteri Delfin (1899b; pp. 119-120) gives 
some color notes for the South American representatives of the species. The color 
of the iris is reddish yellow and the conjunctiva is green speckled with greenish 
yellow spots. The cheeks are described as hoary, with a scaled appearance due to 
the coloration, which probably refers to the dark vermiculations described above, 
which do sometimes look like scales. Most of the body is a greenish brown, 
with blackish overtones above, becoming paler ventrally; there also may be 1 
or 2 longitudinal bands. Rays of pectoral fins with yellow spots, largest basally; 
the axil is yellow. Membranes of dorsal and anal fins dusky green, with spots of 
two shades of greenish yellow. Caudal fin greenish brown with a pale vertical 
band. 

Distribution. I can find no essential differences between the New Zealand 
and South American material and I therefore concur with Norman (1937b; p. 91) 
that specimens from the two areas represent the same species. Such a broad 
distribution is not surprising when one considers the broad distributions of other 
closely related species {N . coriiceps, A^ rossii, and N. magcllanica), all of which 
have characteristic pelagic young. Although no pelagic juveniles of this species 
have been found, it is probable that they do exist. Night-light fishing in the 
waters east of New Zealand might prove fruitful, for many pelagic juveniles of 
the other three species have been obtained in this manner. 



Vol. XXXVIII] DeWITT: NOTOTHENIA FROM NEW ZEALAND 323 

Discussion. The use of the specific name "angustata" for this species and 
the inclusion of Notothenia porteri in its synonyny represents a radical departure 
from the interpretations of all workers since Hutton's time. My reasoning is 
as follows. 

Probably no one took the trouble to read Delfin's description carefully, for 
the number of spines and rays in the dorsal and pectoral fins, and the color 
description, which have been abstracted and brought together above, clearly 
indicate that the species cannot be X. magellanka (Di 4-6; Do 28-30; P 18-19 
in -V. porteri vs. Di 3-6; D- 29-31 ; P 16-18 in .Y. magellanka). 

The realization that something was amiss in the interpretations of Hutton's 
work by later authors came as a result of attempting to place the various early 
names applied to New Zealand nototheniids with the 5 species recognized 
from the area by Parrott (1958). Although it became clear that Hutton had 
himself confused species, certain important discrepancies were found between 
his descriptions of N . angustata and A^. mkrolepidota and the species to which 
the names were applied. These include fin ray counts, scale counts, color, and 
shape of the caudal fin. I concluded that the name Notothenia angustata should 
apply to the species which has been called .V. inter olepidota by all authors since 
Hutton's time, and that the latter name should apply to the species w^hich have 
been called A"", eolbecki and A', jilholi (see discussion under A'', micr olepidota). 
The confusion can probably be traced back to the 1880's, when a number of 
fishes were given to the British Museum by the Otago Museum, including some 
type material. Among these fishes is a specimen identified as N . micr olepidota 
which, although never labeled as such, was presumed to be the type (see Norman, 
1937b; p. 89). I have examined this specimen (BMNH 1886.11.18.30) and it 
does belong with the species described here. Boulenger (1902; p. 185) was the 
first to apply the name .V. microlepidota to the present species and, because he 
gave in the same paper an excellent description of the true N . microlepidota under 
the new name N. eolbecki, all later authors followed him. 

In an effort to obtain further and better evidence to support my belief that 
Hutton's species had been confused, I wrote to the Otago Museum in Dunedin. 
New Zealand. Dr. D. R. Simmon of that institution was kind enough to locate 
the Notothenia material in the museum and to discover that the types of A^ 
angustata and A', microlepidota are probably there. Although there are no data 
which demonstrate that the Otago Museum specimens definitely are the types, 
the circumstantial evidence is very strong. In his 1876 paper, which redescribes 
both A', angustata and A. microlepidota, Hutton stated that the types of both 
species were in the Otago INIuseum. Further, the lengths given by Hutton are 
close to those measured by myself on the stuffed specimens, my measurements 
being greater for both species (A'^. angustata: Hutton's length "about 14.5 
inches," equals 368 mm.; my measurement, TL = 407 mm.; A^. microlepidota: 



324 CALIFORNIA ACADEMY OF SCIENCES IProc. 4x11 Ser. 

Hutton's length "about 17 inches," equals 432 mm.; my measurement 492 mm.). 
Mr. P. O'Brian, the preparator at the Otago Museum, stated that the process of 
stuffing tends to lengthen specimens slightly, and this may account for the ap- 
parent greater size of the stuffed specimens. There are other discrepancies be- 
tween the original accounts and my own data, the most serious being the pectoral 
count of 18 for N. microlepidota (I counted 21 rays). This count is difficult 
to make on large specimens, however, because of thick investing skin, and I can 
only conclude that Hutton's count is in error. The same may be said for my 
scale counts, which were made with difficulty because of the heavy lacquer with 
which the specimens are coated. Other differences between Hutton's published 
accounts and my own observations include dorsal and anal fin ray counts. 

Whether one believes, as I do, that the specimens in the Otago Museum are 
the types, or because of the above discrepancies one believes that they are not, 
the interpretations of Hutton's species by Boulenger, Waite, Norman, and Regan 
are untenable. N otothenia angustata is described as having "a bony ridge over 
each eye extending back to the posterior margin of the praeoperculum," the 
"Caudal rounded," the "Lips speckled with white," 19 rays in the pectoral 
fin and 52-58 scales in a lateral longitudinal series. The supraorbital ridges, 
rounded caudal fin and number of rays in the pectoral fin clearly distinguish it 
from A^. magellanica and demonstrate that it is the same as the A^. microlepidota 
of the above authors. Hutton described N . microlepidota as having 91 scales in 
a lateral longitudinal series, 12 scale rows between the origin of the second dorsal 
fin and the upper lateral line, and a truncate caudal fin. These characters are 
all incompatible with the species which has been called A^. microlepidota and 
show its identity with A^. colbecki and A^ jilholi (see discussion under A^. micro- 
lepidota). Finally, the name ^^ microlepidota'^ certainly refers to the small and 
numerous scales implied by the high counts given in the original description, and 
which is most inappropriate if the conventional interpretation of the 2 species 
is accepted. Table 4 presents the data for the types of A^ angustata and A^. 
microlepidota, together with those for the types of other species synonymized 
with them. 

I have seen the holotype of N. latijrons (USNM 76854), but it is not in 
good condition and I did little with it. Thompson's description of the species 
seems good, and since the upper lateral line count of 51-56 is diagnostic for 
this species I again concur with Regan (1916) and Norman (1937b) that A^. 
latijrons belongs in the synonymy of N . angustata. 

N otothenia parva v/as described from 4 specimens ranging in size from 3 to 
3V2 inches in length (equals 76-89 mm.). I have been able to locate three of 
these specimens, which are now deposited in 3 different institutions: the British 
Museum, the Dominion Museum, and the Otago Museum. Since the specimen in 
the British Museum is not mounted in gelatine or on a glass plate and is the 



Vol. XXXVIII] DeWITT: NOTOTHENIA FROM NEW ZEALAND 325 

most accessible, I designate it as the lectotype, the two others thereby becoming 
paralectotypes. 

The only remaining nomenclatural problem concerning this species is the 
identity of A', maoriensis Haast, which has priority over the name '^angustata." 
However, it is not possible to determine without any doubt whether the original 
description applies to A\ magellanka or to the present species. Characters which 
indicate an identity with A", magellanka are the first dorsal fin with only 3 
spines, the lack of any mention of supraorbital ridges on the head, and the dark 
coloration and lack of any speckling on the head. Characters which indicate an 
identity with A\ angustata are scales present below the eye, the posterior end of 
the upper lateral hne ending below the last ray of the second dorsal fin, and the 
shape of the pectoral fin as shown in the figure published with the original 
description. The illustration might constitute conclusive evidence except that 
it is a relatively crude drawing and contains some obvious errors which indicate 
it was not made with care. In the description the second dorsal fin is said to 
have 29 rays; the drawing shows only 27, and the membranes are drawn in a 
manner not found in any specimens belonging to N otothenia. The pelvic fins 
are drawn with a spine and 6 rays with the first ray longest; I have never ex- 
amined a specimen of N otothenia with 6 pelvic rays, and the third or fourth 
rays are longest, never the first. For these reasons I cannot trust the shape shown 
for the pectoral fin. Further, large specimens of N . magellanka have a number 
of low papillae below and behind the eyes, many of which are broad and flattened 
and appear similar to scales. It is possible that these were mistaken for scales 
by Haast. 

To conclude, there is enough doubt concerning the identity of N . maorknsis 
to make me follow Regan and Norman in placing it with N . magellanka. 

Notothenia microlepidota Hutton. 

Notothenia microlepidota Hutton, 1875: 316 (original description; type locality Dunedin 
and Moeraki (45°23'S., 170°52'E.), New Zealand; holotype in Otago Museum, Dunedin) ; 
Hutton, 1876: 213 (virtual reprint of 1875 original description); Hutton, 1879: 339 
(listed with counts); Hutton, 1890: 280 (listed); Gill, 1893: 118 (listed); Waite, 
1907: 30 (listed); Fowler, 1945: 130 (listed). 

Nototoenia filholi Sauvage, 1880: 228 (original description; type locality Campbell Island; 
types in Museum National d'Histoire Naturelle, Paris); Filhol, 1885: 345 (reprinting 
of Sauvage's description). 

Notothenia filholi Dollo, 1904: 127 (listed); Vaillant, 1907: 22-23 (redescription of syn- 
types; correction of errors made by Sauvage); Regan, 1913: 278 (description from 
Vaillant and Sauvage) ; Phillipps, 1927a: 13 (listed) ; Phillipps, 1927b: 44 (listed) ; 
Blanc AND Hureau, 1962: 341-342 (disposition of syntypes). 

Notothenia colbecki Boulenger, 1902: 185, pi. 16 (original description and illustration; type 
locality Campbell Island; types in British Museum); Waite, 1907: 30 (Hsted) ; Waite, 
1909: 594-595 (description); Regan, 1913: 278 (description); Waite, 1916: 70 (listed); 
Regan, 1916: 378 (distribution); Rendahl, 1925: 6 (listed); Phillipps, 1927a: 13 



326 



CALIFORNIA ACADEMY OF SCIENCES 



IProc. 4tii Ser. 



■i; -s 



s "H 



a I 

■ u 



M g « 



'« ^ 






s ? 



t3 ■« 



U-0i"8'iri06l 
HNKa ■■i!/->3qjoj -jV 



(adXiopaq) 

TZ-0i'8-irt061 

HNIMa -f^MQio:) A' 



(9Cl;t}0pa]BJ'E J ) 

fl-cfS'lVlQ6\ 



t^i-^r8■^■I06I 

HNIMa -mJaqioo -M 



(adX;op3|BJBj ) 
HNIMa ■mo^qioJ -M 



(adX)op9']) 

nnv iM.i 
■nomil A' 



(3dXlOp3[BJEJ) 

fUZV IM.I 



(3dXx) 
:v)op!43]o.uim -^ 



(3dX)BJ'BJ) 

Vl-L-9?6l HNIMa 
:vjnioSvjv<J '^ 



(adXioiOH) 
I-Of-IIX-?I dllM 

■.DJtuoSvjvc} ■ {^ 



(adXjopaq) 

62"8l"n"988I 

HNIMa :nii-""/ A/ 



CsdXx) 
■.ii)nisiiSuv '^ 



uojiEAjasqo 



Cvi liO rv) rvj 
x-^ O f^l O^ 



liO t^ "^ lo O ^O C7^ »-H o^ 00 
rot^rsli-HrOf^^OOOO 



CO --H 



O J~-; M3 CO O 
10 VO O <^ O -^ _ 



*-H Q\ CM 



O ^O CC J^ 



C0^O0^"^^N00lOloC^0^r0•^^O^ 



1— IVOCOIO^^OO^O*— troCOrH 



I--- 00 t^ I— I 

I^ CSl O Tf 

i-Hr-l.— I ogi-Hi— irO'^iO 



10 00rN^OOOOO^^C^Oqrf^v50^0TtU-) 



10 r<i o 

(-0 cy> r-^ 



00 T— ( ro t^ 



1^ r^ 1— I 



(VI rt; -51; 0\ T^ 
x^ O >o "Tf 1~-^ 

T-H T-H CN) >— I 1— ( 



rt•■^^r)C^0^'^O^0(^^0^0•^^-Hr^ 



ro fN »/^ rO (Nl yD 
vd -^ 06 CO r^ v6 
0^ T-i rsi rH th rN 



r^ O "-I O "* 






"^ rt 



C C 



■ O P ^ Oh &H Q 



> Ph 



Q Q < 



ffiOc«w5;Z;yCL,i->UUPqmQHt/5CA!CA! 



Vol. XXXVIII] DeWITT: NOTOTHENIA FROM NEW ZEALAND 



327 



(adX^opaiBJBj) 
U-0r8"Tri06I 



(adAiopaq) 

u-oz-8-iri06i 

HNIMa ■■til33qj03 -^ 



(adXiopa(BJB(j) 
fl-Zf2'\V\06\ 



(adXjopaiBJBj) 



(3dX}Op9[EJBJ) 

HNKa •■!^33q}03 A' 



(adX}opaq) 

i/ge^v Kd 



(adXjopacBJBj) 
t'gfZV Kd 






(adXjBJEj ) 

fZi-Qeei HNKa 

:v3tuoSv)v4 'M 



(adX}0|OH) 
T-Of-IIX-?I dlK 

:v3iuo3vjv<i 'M 



(adXjopaq) 

6Z'8rir98SI 

HNKa :"'!•""/ \V 



:piv}shSud 'A' 



ro yD oo ^ . 

lO ro fO ^^ -|- 

+ 



r^ 1^ ^ '^t 

(vj (^1 >—( og 



■^ CM T^ o^ CO 



^ t^ \o ■* I 
o "^ o lO -|- 

+ 



J>.r^rOTf fOGO»-HO\ 



m3 rO 

;:: 5: ca oa 



10 00 

0^ "^ 



q ^, -H^ 

ON --i 0^ 



CO f^. <~-) 
i-i ON ro 



+ 



o 



r~. r^ ~:) tJ- O 
c^ r-j T-i (^j 



00 1^ <r) 



3 r^i rt CN 



--1 ro ca oa 

'"' ■* ID vo 



cvi CO 

=3 ri 



(v) 3 r-i r<i 



■* O CO ti 











PO 00 


MD 


O^ 


CO 


"* ^ oa ti 

T-H CN) ^^ 3 



-* •^ vq cN 4" 

t-^ ON vd >— I >— I 

vO CO Tj- ■^ I 



.-I Tf 10 "-I 

ly-)ON':fvOoao3<^NOOOoao3 
00 CO 0^ rs) 



^ o3 oa 

t^ 10 vO 
1^ 1— ( 



+ 


ON 








1-H 


n 


vO 


ON 


■* 


+ 




r^ 


CN 



^. CN, NO + - 




■* r^ 10 ,1 

- - - + II 


J--* o^ to 10 


r^j c^i 1— 1 



"^ <^ ^ 



lo .—I 

00 o3 (=a 



oa "^ 



pi; 

> ^ 
■7 hJ hJ i-J c 
< U CL, > < 



^^ Q 9 J kJ 



328 



CALIFORNIA ACADEMY OF SCIENCES 



[Proc. 4th Ser. 



■i^- 



■~t' 





Figure S. Notolhenia microlepidota. Lateral view and top of head of adult, and lateral 
view of young; from Boulenger, 1902. 



(listed); Phillipps, 1927b: 44 (listed); Norman, 1938: 27 (distribution); Parrott, 
1958: 112-113 (description). 

Material Examined. PM A2384: Campbell Island (2; about 126 and 151 

mm., not in good condition; paralectotype and lectotype, respectively, of 

N. jilholi). 
BMNH 1901.11.8.70-71: Campbelllsland (2; 196 and 331 mm.; paralectotype 

and lectotype, respectively, of N . colbecki). 
BMNH 1901.11.8.72-74: Campbell Island (3; 73.1-111 mm.; paralectotypes 

of iV. colbecki). 
DM 2734: Campbelllsland (4; 111 141 mm.). 

The following New Zealand material was also examined, but not used for 
descriptive purposes. 

In the Dominion Museum: 1413, Tucker Cove, Campbell Island (1); 
2084, off Big South Cape Island (1); 3123, off rocks in N. W. Bay, Campbell 
Island (1) ; ?>ii?), outer Ranui Cove, Auckland Island (2). 

In the Canterbury Museum (uncatalogued) : Campbell Island (3); Per- 
severance Harbour, Campbell Island, from throat of Shag (1); Perseverance 
Harbour, Campbell Island ( 1 ) ; Penguin Harbour, Campbell Island ( 1 ) ; Auck- 
land Islands (2). 

Description. Body fusiform, compressed throughout, including head (ex- 
cept in largest specimen) ; dorsal and ventral profiles nearly evenly curved 
throughout, a little more strongly so anteriorly, dorsal profile sometimes slightly 



Vol. XXXVIIIl DeWITT: NOTOTHENIA FROM NEW ZEALAND 329 

more convex than ventral profile; caudal peduncle distinctly longer than deep. 
Length of head 288-341, its width 126-247, its depth 164-182; depth of body 
182-232, its width 115-190, pectoral to pectoral distance 138-254, dorsal to 
anal distance 196-254; length of caudal peduncle 123-137, its depth 79-91; 
dorsal to caudal distance 124-144. Vertebrae 18 + 27-28 = 45-46. 

Snout smoothly rounded from both lateral and dorsal views, rising from 
tip of upper jaw at about same angle as top of head; its length 81-100. Nostrils 
tubular, elliptic in cross section, each with its hind margin raised into a flap 
ending in a rounded point; nostrils placed 48-69 from tip of snout, 17-23 from 
orbit, and 52-63 apart. 

Eyes directed laterally, placed high on head, above a line between tip of 
snout and upper end of base of pectoral fin, but not protruding into dorsal profile 
of head; diam.eter of orbit 51-77. Interorbital space broad and nearly flat, only 
very slightly convex, its width 66-103. Length of postorbital part of head 
154-185. 

Mouth oblique, lower jaw projecting slightly in front of upper jaw; length 
of upper jaw 104-128, maxillary extending posteriorly under first third of eye. 
Teeth in upper jaw may be described for convenience as being in 2 bands; outer 
band a uniserial row of enlarged, spaced (canine-like) teeth, extending only along 
anterior half of jaw; inner band composed of smaller, more closely spaced teeth, 
slightly broadened anteriorly, becoming a uniserial row posteriorly; inner teeth 
become slightly larger posterior to point where outer row ends. Teeth in lower 
jaw may be described as occurring in a single band, somewhat broadened an- 
teriorly, with outermost teeth largest, and becoming a uniserial row of enlarged 
teeth in posterior two-thirds of jaw, the teeth becoming smaller posteriorly. Oral 
valves extend nearly entire length of jaws; they may be covered with papillae or 
not. Tongue rounded and free anteriorly, with a slight depression in its upper 
surface, and covered with scattered low papillae. 

Gill rakers in anterior series of first gill arch slender and elongate, arranged 
6-11 + 0-1 + 15-19 = 24-30; those on lower limb near angle slightly flattened 
on ventral edge, those further below flattened dorsoventrally, those on upper 
limb more cylindrical; all bear a few to many teeth, those on lower limb near 
angle with fewest. Posterior gill rakers of first arch short and blunt, somewhat 
flattened dorsoventrally, and bearing teeth; arranged 1-3 + 1 + 14-15 = 17-19. 
Branchiostegal rays 6. 

First dorsal fin 6-8, its origin 295-338 from tip of snout, from above upper 
end to just in advance of bases of pectoral fins; its height relatively low, length 
of longest spine 86-110. Second dorsal fin 25-29, its origin 413-467 from tip 
of snout, 28-48 from base of last spine of first dorsal fin; highest anteriorly, 
length of sixth ray 105-130, of sixth from last ray 75-79. Anal fin 21-24, its 
origin 507-570 from tip of snout, below bases of rays five to seven of second dorsal 



330 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

fin; highest anteriorly, length of sixth ray 88-111, of sixth from last ray 76-87. 
Caudal fin 14, its length 173-216, its posterior margin distinctly emarginate, 
almost forked. Although the sample counted is small, the apparent lack of 
variation in the number of principal rays may be due to the emarginate shape 
of the fin, in which the principal unbranched rays are nearly as long as the 
longest branched rays and form most of the upper and lower edges of the fin. 

Pectoral fins 20-21, their length 190-232, not reaching to, or extending as 
far as, above fourth ray of anal fin, the posterior margin rounded; width of their 
bases 66-88. Pelvic fins placed 251-314 from origin of anal fin, entirely in 
advance of bases of pectoral fins; their length 156-197, third ray longest, not 
reaching posteriorly to origin of anal fin. 

Upper lateral line with 61-75 tubular scales, ending below last few rays of 
second dorsal fin or extending a short distance posterior to it, separated from 
origin of second dorsal fin by 9-11 scale rows. Boulenger (1902; p. 185) gives 
a range of 59-71, but his counts are low in every case. Table 4 presents my 
counts from the same specimens (see under discussion), which range from 67-75. 
Middle lateral line with 24-37 tubular scales, originating below ninth to fifteenth 
rays of second dorsal fin, and extending onto base of caudal fin. 

Cephalic lateral-line canals normal in pattern except that preoperculo- 
mandibular canals are joined to temporal canals. The pores are small and diffi- 
cult to find. Preoperculo-mandibular canals with 9-10 (usually 9) pores, 
connected to temporal canals at areas of second pores of latter canals; infra- 
orbital canals with 9-11 (usually 10) pores; supraorbital canals each with 4 
pores and sharing a median coronal pore; temporal canals with 6 pores; supra- 
temporal canal with 2-4 (usually 3) pores. 

Scales everywhere small, 84-98 in a lateral longitudinal series, with 37-45 
rows around caudal peduncle; on body ctenoid except dorsally anterior to first 
dorsal fin, anterior to bases of pectoral fins, on ventral surface anterior to pelvic 
fins, and sometimes on lower sides of body between pelvic fins and anterior few 
rays of anal fin. A few nonctenoid scales may be found scattered among the 
ctenoid scales, especially at base of caudal fin, and the number of ctenae may 
be reduced to one. Scales extend onto basal parts of caudal fin and, except for 
a naked arc at bases of rays, onto exposed proximal portions of pectoral fins. 

Most of head naked; 2 patches of scales, some of which are ctenoid, present 
on each side at posterolateral corners of head, one just anterior to supratemporal 
canal, the other in triangle formed by temporal canal, supratemporal canal and 
very weak ridge of posttemporal bone. An elongate patch of nonctenoid scales 
present on uppermost part of operculum; a patch of similar scales present on 
upper and anterior part of cheek, extending ventrally and anteriorly in a nar- 
rowing arc around posterior and ventral margins of eye. Upper portions of head, 
including snout, lips, and naked parts of cheeks, as well as other parts in lesser 



Vol. XXXVIII] DeWITT: NOTOTHENIA FROM NEW ZEALAND 331 

degree, covered with scattered and low papillae or ridges, the most marked being 
on top of head and anterior parts of lower lip and jaw. 

Ground color of body (in alcohol) uniformly brownish or greyish, becoming 
lighter ventrally; lower half of body may be somewhat silvery (this probably re- 
flects the method of initial fixation). Both dorsal fins dusky to deep brown; 
anal fin pale to brown; pectoral fins slightly brownish basally; pelvic fins a 
little dusky distally; caudal fin slightly dusky. Upper surface of head and 
tip of lower jaw dark, head otherwise becoming lighter ventrally; lower halves 
of operculum and cheek may be silvery. Indistinct and irregular dark areas 
may be present on top of head; a dark patch may be present behind eye at 
level of upper end of preopercular. Two dark lines may be present on lower 
parts of cheek, one extending from edge of upper jaw above end of maxillary 
posteriorly and ventrally towards lower margin of preopercular, the other ex- 
tending from ventral margin of eye towards angle of preopercular. A third 
line, extending from posteroventral edge of eye to upper end of preopercular, 
may be present, and the dark patch behind the eye mentioned above may repre- 
sent this line. 

Juvenile specimens are somewhat silvery in color and, although there are 
no striking color changes between the young and adults as seen in N. rossii, 
the silvery color may indicate that the young of this species are also pelagic 
in habit. 

Little has been recorded of life colors. Hutton (1875) gives "Purplish 
brown above, greyish below; throat, gill-membranes, axil of pectorals, and 
opercles yellowish." Parrott (1958) notes that a specimen from Auckland 
Island was dark olive-green with dark red bands on the dorsal and ventral fins. 

Distribution. Notothenia microlepidota is known only from the New 
Zealand region, including Macquarie Island. Its habits are apparently similar 
to those of N . angustata, specimens having been captured primarily with hooks 
and lines close to shore. 

Discussion. Since the nomenclature and synonymy used for this species 
are totally different from those used by previous authors, some explanation of 
the present usage is desirable. My first suspicion that the names A' . filholi and 
TV. colbecki represented the same species was entertained upon reading the 
account by Filhol (1885; pp. 343-346) of his fishing efforts at Campbell 
Island. He stated that A^. jilholi was the most common fish encountered there. 
Boulenger later described A. colbecki, also from Campbell Island, and it sub- 
sequently was found to be very common there, whereas A. jilholi was never 
recorded again. Boulenger's description is good and it was accompanied by 
an excellent figure; A. colbecki was therefore easily recognized by subsequent 
workers. Sauvage's description, on the other hand, is not only brief, but con- 



332 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

tains a number of important errors (see Vaillant, 1907; p. 22, footnote), and 
no illustration was prepared. Thus N. jilholi was never again recognized, al- 
though the name continued to be included in keys and check lists because of 
the unusual counts which Sauvage had given. Vaillant (1907) corrected 
Sauvage's errors, but his redescription and discussion has been disregarded. 
Regan (1913), apparently not knowing what to believe, gave the data of both 
Sauvage and Vaillant; later authors followed Sauvage. 

Through the courtesies of Dr. Maurice Blanc of the Museum National 
d'Histoire Naturelle, Paris, and of Mr. A. C. Wheeler and Dr. P. H. Green- 
wood of the British Museum (Natural History) I have been able to examine 
2 syntypes of N . jilholi and 5 syntypes of N . colbecki. Although the specimens 
of A^. jilholi are not in good condition, I was able to take some counts and 
measurements from them. These are presented in table 4 together with the 
more complete data from the specimens of N . colbecki. There is no doubt that 
they all represent the same species, and the data from them have been incor- 
porated into the above description. 

I have related already the probable cause of the confusion attending Hutton's 
species A^. angustata and N. microlepidota (see discussion section under A^. 
angustata). Although there is some doubt whether the specimen in the Otago 
Museum thought to be the type of N . microlepidota is indeed the type, since 
there are no records or catalogues dating back to the 1870's, the original de- 
scription leaves no doubt that Hutton's species is the same as both N . jilholi 
and N . colbecki. The supposed type is now stuffed, and while its total length 
is somewhat greater than that recorded by Hutton, it is sufficiently near 
Hutton's figure that the difference can be accounted for by the process of 
stuffing. The counts taken from the specimen are presented in table 4 for 
direct comparison with those from the types of A^. jilholi and A^. colbecki, and 
show conclusively that the specimen, whether type or not, represents the same 
species as the others. 

A final matter is the selection of lectotypes from the type series of N. 
jilholi and A^. colbecki. As the lectotype of the former name I choose the 
specimen of 151 mm. standard length (Paris Museum number A2384) listed 
above in the material examined. This is as nearly as I can tell the specimen 
which Vaillant used for his table (1907; p. 23) and is probably the specimen 
referred to by Sauvage in his original description when he gave a length of 
350 mm. (corrected to 150 mm. by Vaillant, 1907; see also Blanc and Hureau, 
1962; pp. 341-342). P'or the lectotype of the name A^. colbecki I choose the 
specimen of 331 mm. standard length (British Museum number 1901.11.8.70-71) 
listed above in the material examined. This is the largest of the specimens 
which remain of the type series, and is probably the specimen used for the 
figure of the adult published with the original description (there is some 
doubt because the legend for the plate states that the figure has been reduced 



Vol. XXXVIII] DeWITT: NOTOTHENIA FROM NEW ZEALAND 333 

to V3, which would mean the specimen was a Httle over 500 mm. in standard 
length; if the above specimen was used for the illustration the reduction is 
about V2). Boulenger also gives in his description a total length of 380 mm., 
which corresponds well with both Norman's (1938; p. 27) and my measure- 
ments (385 and 388 mm., respectively) for the largest specimen in the series, 
and indicates that this specimen was considered as the type. Only 5 of the 
original 12 specimens of the type series remain, and they are undoubtedly the 
5 specimens Boulenger used for his table of counts and measurements (Boulen- 
ger's total lengths: 380, 230, 130, 120 and 85; my measurements: 388, 232, 
130, 121 and 89). 

ACKNOWLEDGMENTS 

This paper is an extension of a portion of a doctoral dissertation submitted 
to the Department of Biological Sciences of Stanford University, and I take 
this opportunity to express my appreciation to Professor George S. Myers 
who, as my advisor, greatly aided me in formulating my dissertation topic 
and guided me through the initial stages of the work. The late Miss Margaret 
Storey of Stanford was very helpful in searching the literature and often gave 
me needed advice and encouragement. Much of the work was done at the 
University of Southern California where I participated in the Antarctic Marine 
Biological Research Program there, supported by the Antarctic Office of the 
National Science Foundation (grants G 19497, GA 238 and GA 448). Travel 
to and from the Antarctic research vessel Eltanin under these grants enabled 
me to visit museums in Argentina, New Zealand and Australia. Travel grants 
(GA 180 and GA 376) from the Antarctic Office enabled me to visit the 
British Museum in 1965 where I was able to examine the extensive Antarctic 
fish collection there and the Zoological Institute in Leningrad in 1966 where 
I was able to examine some of the recently collected Russian Antarctic material. 
I am much indebted to Dr. Jay ^L Savage for leadership and advice during 
this period. The final writing was accomplished at the University of South 
Florida, St. Petersburg Campus. 

For the loan of specimens or for the innumerable courtesies shown me 
during my visits to museums in this and other countries, I give my most sincere 
thanks to the following persons: Drs. L. P. Schultz, E. A. Lachner, V. G. Springer, 
and S. H. Weitzman, and to Mr. R. Kanazawa of the United States National 
Museum; Dr. P. H. Greenwood, Mr. A. C. Wheeler, Dr. N. B. Marshall, and 
Miss V. duHeaume of the British Museum (Natural History); Drs. A. N. 
Svetovidov and A. P. Andriashev of the Zoological Institute in Leningrad; 
Professor K. Deckert and Dr. C. Karrer of the Zoologisches Museum, Humboldt- 
Universitat, Berlin; Professor M. Blanc of the Museum National d'Histoire 
Naturelle, Paris; Dr. P. Kahsbauer of the Naturhistorisches Museum, Vienna; 
Drs. R. Lopez and N. Bellisio of the Museo Argentino de Ciencias Naturales, 



334 CALIFORNIA ACADEMY OF SCIENCES IProc. 4th Ser. 

Buenos Aires; Dr. R. A. Ringuelet of the Museo de La Plata, La Plata; Mr. 
John Moreland of the Dominion Museum, Wellington; Mrs. Marie Darby 
(nee Biichler) of the Canterbury Museum, Christchurch; Dr. D. R. Simmon 
of the Otago Museum, Dunedin; and Mr. C. J. M. Glover of the South Australian 
Museum, Adelaide. 

I also thank the Trustees of the British Museum (Natural History) for 
permission to publish a reproduction of J. G. A. Forster's manuscript illustration 
of Gadus magellanicus (fig. 1). Figure 3 was made possible through the 
courtesy of Dr. D. M. Cohen of the United States Bureau of Commercial 
Fisheries Ichthyological Laboratory, Washington, D. C. The staff in the 
Reference Department of the main library at the University of South Florida 
were very generous in their efforts to obtain for me publications not in their 
holdings. 

Finally, I thank Dr. Thomas L. Hopkins for critically reading portions of 
the manuscript. 

LITERATURE CITED 

Andriashev, a. p. 

1959. On the number of vertebrae and some osteological characteristics of Antarctic 
fishes as shown by X-ray photographs. Voprosi Ikhtiologii, Moskva, no. 12, 
pp. 3-7 (in Russian). 

1965. A general review of the Antarctic fish fauna. Monographiac Biologicae, vol. 15, 

pp. 491-550. 

Andriashev, A. P., and A. K. Tokarev. 

1958. Ichthyofauna. The study of ichthyofauna and the goals of investigations. Trudy 
kompleksnoi antarkticheskoi Ekspeditsii. Akademiia Nauk S. S. S. R. (Opisanie 
Ekspeditsii d/e "Ob" 1955-1956), pp. 195-207 (in Russian). 

Bellisio, N. B. 

1966. Fauna Marina Antartica. Republica Argentina, Secretaria de Marina, Servicio 

Hidrografia Naval, Publico H. 907, 93 pp. 
Blanc, M. 

1951. Poissons recueilhs aux lies Kerguelen par le Docteur Aretas. Bulletin du Museum 

National d'Histoire Naturelle, Paris, ser. 2, vol. 23, pp. 493-496. 
1954. Poissons recueillis aux lies Kerguelen par P. Paulian (1951) et M. Angot (1952). 

Bulletin du Museum National d'Histoire Naturelle, Paris, ser. 2, vol. 26, pp. 

190-193. 
1958. Sur quelques poissons des lies Kerguelen rapportes par le Dr. Bourland. Bulletin 

du Museum National d'Histoire Naturelle, ser. 2, vol. 30, pp. 134-138. 

1961. Les poissons des terres australes et antarctiques fran(;aises. Memoires de 

rinstitut Scientifique de Madagascar, ser. F, vol. 4, pp. 109-159. 
Blanc, M., and J.-C. Hureau. 

1962. Catalogue des types de poissons de la famille des Nototheniidae, en collection 

au Museum National d'Histoire Naturelle. Bulletin du Museum National 
d'Histoire Naturelle, Paris, ser. 2, vol. 34, pp. 339-342. 

BOULENGER, G. A. 

1900. A list of the fishes collected by Mr. Rupert Vallentin in the Falkland Islands. 
The Annals and Magazine of Natural History, ser. 7, vol. 6, pp. 52-54. 



Vol. XXXVIII] DeWITT: NOTOTHENIA FROM NEW ZEALAND 335 

1902. Pisces, pp. 174-189, pis. 11-18. In Report on the Collections of Natural History 
made in the Antarctic Regions during the Voyage of the Southern Cross, 
London, British Museum (Natural History). 
Cunningham, R. O. 

1871. Notes on the Reptiles, .-Xniphibia, Fishes, Mollusca, and Crustacea obtained 
during the voyage of H.M.S. Nassau in the years 1866-69. Transactions of 
the Linnean Society of London, vol. 27, pp. 465-502. 
Delfin, F. T. 

1899a. Catalogo de los peces de Chile (continuacion). Revista Chilena de Historia 

Natural, vol. 3, pp. 15-24. 
1899b. Descripcion de un nuevo Traquinido Chileno. Revista Chilena de Historia 
Natural, vol. 3, pp. 117-120. 
DeWitt, H. H. 

1962. A new Antarctic nototheniid fish with notes on two recently described notothenii- 
forms. Copeia, 1962, pp. 826-833. 
DOLLO, L. 

1904. Poissons. Resultats du Voyage du S.Y. Belgica en 1897-1898-1899 sous le 
commandement de A. de Gelache de Gomery, Rapports Scientifiques, Zoologie, 
Anvers, 240 pp., 12 pis. 
FlLHOL, H. 

1885. Poissons, pp. 343-346. In H. Filhol, Mission de I'lle Campbell: Recherches 

zoologiques, botaniques et geologiques a I'lle Campbell en Nouvelle Zelande. 

Recueil de Memoirs, Rapports et Documents relatifs a I'observation du 

Passage de Venus sur le Soleil, vol. 3, pt. 2, no. 1. 
Fischer, J. G. 

1885. Ichthyologische und herpetologische Bemerkungen. I. tJber Fische von Siid- 

Georgien. Jahrbuch der Hamburgischen Wissenschaftlichen Anstalten, Hamburg, 

vol. 2, pp. 49-65, pis. 1-2, figs. 1-4, 9. 
Forster, J. R. 

1801. Gadus magellanicus, p. 10. In M. E. Bloch and J. G. Schneider, Systema Ichthyo- 

logiae Iconibus CX Illustratum, Berolini. 
1844. Descriptiones animalium quae in itinere ad maris australis terras per annos 

1772 1773 et 1774 suscepto collegit obsevavit et delineavit . . . nunc demum 

editae auctoritate et impensis . . . Henrico Lichtenstein. Berlin, xiii + 424 

+ 1 pp. 
Fowler, H. W. 

1926. Fishes from Florida, Brazil, Bolivia, Argentina, and Chile. Proceedings of the 

Academy of Natural Sciences of Philadelphia, vol. 78, pp. 249-285. 
1945. Fishes of Chile. Systematic Catalog. El Imparcial, Santiago de Chile, 36 + 171 pp. 
1951. Analysis of the fishes of Chile. Revista Chilena de Historia Natural, vol. 51-53, 

pp. 263-326. 
Frost, G. A. 

1928. A comparative study of the otoliths of the neopterygian fishes. The Annals 

and Magazine of Natural History, ser. 10, vol. 1, pp. 451-456, 1 pi. 
Gill, T. N. 

1862. Synopsis of the Notothenioids. Proceedings of the Academy of Natural Sciences 

of Philadelphia, 1861, pp. 512-522. 
1893. A comparison of antipodal faunas. Memoirs of the National Academy of 

Sciences, Washington, vol. 6, pp. 89-124. 



336 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

GiJNTHER, A. 

1860. Catalogue of the Acanthopterygian fishes in the collection of the British Museum. 

vol. 2. London, British Museum, xxi -\- 548 pp. 
1881. .An account of the zoological collections made during the survey of H.M.S. 
Alert in the Straits of Magellan and on the coast of Patagonia. III. Reptiles, 
Batrachians, and Fishes. Proceedings of the Zoological Society of London, 
1881, pp. 18-22, 2 pis. 
Haast, J. 

1873. Notes on some undescribed fishes of New Zealand. Transactions and Pro- 
ceedings of the New Zealand Institute, vol. 5, pp. 272-278, pi. 16. 
Hart, T. J. 

1946. Report on trawling surveys on the Patagonian continental shelf, compiled 
mainly from manuscripts left by the late E. R. Gunther, M. A. Discovery 
Reports, vol. 23, pp. 223-408, 1 pi. 
HUSSAKOF, L. 

1914. Notes on a small collection of fishes from Patagonia and Tierra del Fuego. 
Bulletin of the American Museum of Natural History, vol. ii, pp. 85-94. 
HUTTON, F. W. 

1872. Catalogue with diagnoses of the species, pp. vii-xvi + 1^93. /» F. W. Hutton 

and J. Hector, Fishes of New Zealand. Colonial Museum and Geological 
Survey Department, Wellington. 

1873. Contributions to the ichthyology of New Zealand. Transactions and Proceedings 

of the New Zealand Institute, vol. 5, pp. 259-272, pis. 7-12, 15. 

1875. Descriptions of new species of New Zealand fish. The Annals and Magazine of 

Natural History, vol. 16, pp. 313-317. 

1876. Contributions to the ichthyology of New Zealand. Transactions and Proceed- 

ings of the New Zealand Institute, vol. 8, pp. 209-218. 
1879. Notes on a collection from the Auckland Islands and Campbell Island. Trans- 
actions and Proceedings of the New Zealand Institute, vol. 11, pp. 337-343. 
1890. List of the New Zealand fishes. Transactions and Proceedings of the New 
Zealand Institute, vol. 22, pp. 275-285. 
Kenny, R., and N. Haysom. 

1962. Ecology of rocky shore organisms at Macquarie Island. Pacific Science, vol. 16, 
pp. 245-263. 
Ladiges, W., G. von Wahlert, and E. Mohr. 

1958. Die Typen und Typoide der Fischsammlung des Hamburgischen Zoologischen 
Staatsinstituts und Zoologischen Museums. Mitteilungen aus dem Ham- 
burgischen Zoologischen Museum und Institut, vol. 56, pp. 155-167. 
LONNBERG, E. 

1905. The fishes of the Swedish South Polar Expedition. Wissenschaftliche Ergebnisse 

der Schwedischen Siidpolar-Expedition 1901-1903, vol. 5, pt. 6, 69 pp., 5 pis. 

1906. Contributions to the fauna of South Georgia. I. Taxonomic and biological 

notes on vertebrates. Kungliga Svenska Veten.skapsakadcmicns Handlingar, 
new series, vol. 40, no. 5, 104 pp., 12 pis. 

1907. Fische. Ergebnisse Hamburger Magalhaensischen Sammelreisc, 1892/93, vol. 8, 

16 pp., 1 pi. 
MacDonagh, E. 

1931. Nota preliminar sobre Bovichthys argentimis y Notothenia patagonica n. .spp. 
Notas Preliminares del Museo de La Plata, vol. 1, pp. 99-100. 



Vol. XXXVIII] DeWITT: NOTOTHENIA FROM NEW ZEALAND 337 

1934. Nuevos conceptos sobre la distribucion geografica de los peces Argentines. 

Revista del Museo de La Plata, vol. 34, pp. 21-170. 
1936. Sobre algunos peces marines. Notas del Museo de La Plata, Buenos Aires, vol. 
1, Zoologia, no. 4, pp. 423-429. 
MacDonagh, E., and M. R. Covas. 

1944. Peces Patagonicos y Fueguinos coleccionados por el Doctor Federico G. Lynch. 
Notas del Museo de La Plata, Buenos Aires, vol. 9, Zoologia, no. 76, pp. 
229-241, 2 pis. 
Moreland, J. 

1957. Report on the fishes, p. 34. In G. A. Knox, General account of the Chatham 

Islands 1954 Expedition. New Zealand Department of Scientific and Industrial 
Research, Bulletin 122. 
Norman, J. R. 

1937a. Fishes. Banzare Reports, Ser. B (Zoology and Botany), vol. 2, pp. 49-88. 
1937b. Coast fishes. Part II. The Patagonian region. Discovery Reports, vol. 16, pp. 

1-lSO, 5 pis. 
1938. Coast fishes. Part III. The Antarctic Zone. Discovery Reports, vol. 18, pp. 
1-104, 1 pi. 
Nybelin, O. 

1947. Antarctic fishes. Scientific Results of the Norwegian Antarctic Expeditions 

1927-1928 et Sqq., no. 26, 76 pp., 6 pis. 
1951. Subantarctic and Antarctic fishes. Scientific Results of the "Brategg" Expedition 
1947-48, no. 2, 32 pp. 
Oliver Schneider, C. 

1943. Catalogo de los peces marinos del literal de Concepcion y Arauco. Boletin de 
la Sociedad de Biologia de Concepcion, Chile, vol. 17, pp. 75-126. 
Olsen, S. 

1954. "South Georgian Cod" {Notothenia rossii marmorata Fischer). Norsk Hvalfangst- 

tidende (The Norwegian Whahng Gazette), vol. 43, pp. 373-382. 

1955. A contribution to the systematics and biology of Chaenichthyid fishes from South 

Georgia. Nytt magazin for Zoologi, vol. 3, pp. 79-93. 
Parrott, a. W. 

1958. Fishes from the Auckland and Campbell Islands. Records of the Dominion 

Museum, Wellington, vol. 3, pp. 109-119. 
Perugia, A. 

1891. Appunti sopra alcuni pesci Sud-Americani conservati nel Museo Civico di Storia 

Naturalle di Genova. Annali del Museo Civico di Storia Naturali di Geneva, 

ser. 2, vol. 10, pp. 604-657. 
Peters, W. C. H. 

1876. tjbersicht der wahrend der von 1874 bis 1876 unter dem commando des Hrn. 

Capitan z. S. Freihern von Schleinitz ausgefiihrten Reise S. M. S. Gazelle 

gesammelten und von der Kaiserlichen Admiralitat der Koniglichen Akademie 

der Wissenschaften iibersandten Fische. Monatsbericht der Koniglichen 

Akademie der Wissenschaften zu Berlin, Sitzung der physikalish-mathematischen 

Klasse, 1876, pp. 831-854. 
Phillipps, W. J. 

1921. Notes on the edible fishes of New Zealand, with a record of fishes exposed for 

sale in Wellington during 1918. The New Zealand Journal of Science and 

Technology, vol. 4, pp. 114-125. 



338 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

1927a. A check list of the fishes of New Zealand. Journal of the Pan-Pacific Research 

Institute, vol. 2, pp. 9-15. 
1927b. Bibliography of New Zealand Fishes. New Zealand Marine Department, Fisheries 
Bulletin no. 1, 68 pp. 
Regan, C. T. 

1913. The Antarctic fishes of the Scottish National Antarctic Expedition. Transactions 

of the Royal Society of Edinburgh, vol. 49, pp. 229-292, 11 pis. 
1916. Bibliographical notice. Antarctic and Subantarctic fishes. The Annals and 
Magazine of Natural History, ser. 8, vol. 18, pp. 377-379. 
Rendahl, H. 

1925. Papers from Dr. Th. Mortensen's Pacific Expedition 1914-16. XXX. Fishes 
from New Zealand and the Auckland-Campbell Islands. Videnskabelige 
Meddelser fra Dansk Naturhistorisk Forening i Kobenhaven, vol. 81, pp. 1-14. 
Richardson, J. 

1844-1848. Ichthyology of the voyage of H.M.S. Erebus and Terror, under the com- 
mand of Captain Sir James Clark Ross, R.N., F.R.S., viii -f 139 pp., 60 pis. 
In J. Richardson and J. E. Gray (eds.), The Zoology of the Voyage of H.M.S. 
Erebus and Terror, under the command of Captain Sir James Clark Ross, 
R.N., F.R.S., during the years 1839-1843, vol. 2. 
RuuD, J. T. 

1954. Vertebrates without erythrocytes and blood pigment. Nature, London, vol. 
173, pp. 848-850. 
Sauvage, H. E. 

1880. Notice sur quelques poissons de I'lle Campbell et de ITndo-Chine. Bulletin de 
la Societe Philomathique de Paris, vol. 4, pp. 228-233. 
Slack-Smith, R. J. 

1962. A small collection of fish from Macquarie Island. Memoirs of the National 
Museum of Victoria, Melbourne, no. 25, pp. 13-16. 
Smitt, F. a. 

1897. Poissons de I'Expedition Scientifique a la Terre de Feu. I. Nototheniae. Bihang 

Till Kongl. Svenska Vetenskaps-Academiens, Handhngar, Stockholm, vol. 
23, pt. 4, no. 3, 37 pp., 3 pis. 
Steindachner, F. 

1875. Ichthyologische Beitrage (iii). Sitzungsberichte der K. Akademie der Wissen- 
schaften in Wien, I abt., vol. 72, pp. 29-96, 8 pis. 

1898. Die Fische der Sammlung Plate. Zoologische Jahrbiicher, Supplement-Band 4, 

Fauna Chilensis, vol. 1, pp. 281-338, pis. 15-21. 
1903. Die Fische der Sammlung Plate (Nachtrag). Zoologische Jahrbiicher, Sup- 
plement-Band 6, Fauna Chilensis, vol. 3, pt. 1, pp. 201-214. 
Studer, T. 

1879. Die Fauna von Kerguelensland. Archiv flir Naturgeschichte, vol. 45, pp. 114-141. 
Thompson, W. F. 

1916. Fishes collected by the United States Bureau of Fisheries steamer Albatross 
during 1888, between Montevideo, Uruguay, and Tome, Chile, on the voyage 
through the Straits of Magellan. Proceedings of the United States National 
Museum, vol. 50, pp. 401-476, pis. 2-6. 
Thomson, G. M., and T. Anderton. 

1921. History of the Portobello Marine Fish-Hatchery and Biological Station. Dominion 
of New Zealand, Board of Science and Art, Bulletin 2, 131 pp. 



Vol. XXXVIII] DeWITT: NOTOTHENIA FROM NEW ZEALAND 339 

Vaillant, L. 

1888. Poissons. Mission Scientifique du Cap Horn, 1882-1883, vol. 6 (Zoologie), 

pt. 1, 3S pp., 4 pis. 
1907. Poissons. Expedition Antarctique Franqaise (1903-1905), Sciences Naturelles, 

Documents Scientifiques, Paris, 51 pp. 
Waite, E. R. 

1907. A basic list of the fishes of New Zealand. Records of the Canterbury Museum, 

vol. 1, pp. 3-39. 
1909. Vertebrata of the Subantarctic islands of New Zealand. Pisces, pp. 585-598. 

In C. Chilton (ed.), The Subantarctic Islands of New Zealand, Philosophical 

Institute of Canterbury, Wellington, vol. 2. 
1916. Fishes. Australasian Antarctic Expedition, 1911-1914, Scientific Reports, Ser. 

C — Zoology and Botany, vol. 3, pt. 1, pp. 3-92, 5 pis., 2 maps. 



PROCEEDINGS 

OF THE 

CALIFORNIA ACADEMY OF SCIENCES 

FOURTH SERIES 



Festschrift for George Sprague Myers 



Vol. XXXVIII, No. 17, pp. 341-346; 1 fig. December 31, 1970 



HOW MANY RECENT FISHES ARE THERE? 

By 

Daniel M. Cohen 

Bureau of Coninierical Fisheries 
Washington, /). C. 

It is a pleasure to dedicate this paper to Professor George S. Myers on the 
occasion of his 65th birthday. His interests in ichthyology have ranged widely 
and the topic of this paper seems especially appropriate, not least because he has 
been interested in this particular problem himself. 

Estimates of the number of species of Recent fishes in the current ichthyo- 
logical hterature range from a low of 15,000-17,000 to a high of 40,000. 
Presented below is a brief list of some. 

SOME PAST ESTIMATES OF NUMBER OF RECENT FISH SPECIES 

Bailey (1960) gave 15,000 to 17,000, of which about 45 are Agnatha and 
about 575 are Chondrichthys. His estimate was apparently based on a group 
approach. 

Marshall (1965) mentioned that, "We know more than twenty thousand 
living kinds, but our inventory is by no means complete." He gave no basis for 
his estimate. 

Norman (1963) gave 25,000, with no mention of how the figure was reached. 

]Myers (1958) stated there are, "'. . . .33,000 or more living species of teleosts." 
No mention of method of estimation was given. 

Schultz and Stern (1948) gave a figure of 40.000; however, Schultz (1965) 
later lowered his estimate to 32,000. No basis was given for either figure. 

[341] 



342 



CALIFORNIA ACADEMY OF SCIENCES 



[Proc. 4th Ser. 




Figure 1. Percentages of Recent fish species living in various habitats. 



MY ESTIMATE OF NUMBER OF RECENT FISH SPECIES 

The wide range of figures suggests that a rational estimate, as opposed to an 
educated guess, is difficult for any one ichthyologist. With this in mind I com- 
piled a list of fish families and began to solicit estimates from specialists and to 
consult recent revisions. Seven years have gone by since the initiation of the 
project, and this seems an appropriate time and place to present the results of my 
canvas. For the several groups for which neither colleagues nor recent revisions 
could supply information, I was obliged to consult several large faunal works and 
interpret the results in what I hope was a judicious manner. 

Estimates are intended to be of the number of living species rather than 
described ones. Although approximately 75 to 100 species of Recent bony fishes 
are described each year (Zoological Record), we lack comparable information on 
how many species are placed in synonymy annually. 

The final results of the present survey are: Agnatha about 50; Chondrich- 
thyes 515 to 555; Osteichthyes 19,135 to 20,980. The figures given for bony 



Vol. XXXVIII] COHEN: RECENT FISHES 343 

fishes are two minimums rather than a maximum and a minimum. Most special- 
ists who volunteered a single figure gave it as a minimum. Many colleagues, 
however, gave a range. The first figure, 19,135, is the sum of single estimates 
and the lower figures of ranges; therefore, it represents a bare minimum. The 
second figure is the sum of single estimates and the upper figures of ranges; 
therefore, it is a combination of minimum and maximum estimates. 

I have attempted an ecological analysis of the data for Osteichthyes. The 
figures used for calculating percentages are averages of high and low estimates. 

(1) Primary freshwater (Myers, 1949) 6650. 33.1 percent. Approximately 
6200 of this group belong to the Ostariophysi. 

(2) Secondary freshwater (Myers, 1949) 1625. 8.1 percent. INIost of the 
species in this group belong to the families Cichlidae, Cyprinodontidae, and 
Poeciliidae. 

Total freshwater 8,275. 41.2 percent. If this astonishingly high percentage is 
valid it must be a reflection of the degree of isolation possible in the freshwater 
environment. 

(3) Diadromous (including Complementary of Myers, 1949) 115. 0.6 per- 
cent. As the systematics and life histories of tropical shore fishes become better 
known it seems likely that at least some species will be shifted from category 4 to 
this group. 

(4) Marine shore and continental shelf to depths of approximately 200 me- 
ters — warm water 8000. 39.9 percent. Perciform fishes and their derivatives are 
the major component of this category. Particularly important are percoid, blen- 
nioid, and gobioid fishes. Among nonperciforms, eels probably contribute the 
most species. 

(5) Marine shore and continental shelf to depths of approximately 200 me- 
ters — cold water 1130. 5.6 percent. A factor that may contribute to the sub- 
stantially smaller size of this fauna as compared with that of group 4 is the 
smaller area occupied. Also, much of the region has had long-term, unstable 
climatic conditions so that many of the species must be fairly recent in their 
present habitats. There is no doubt, however, that a high degree of endemism 
prevails. Important components of this group are Gadidae, Zoarcidae, northern 
blennioids, and scorpaeniform fishes. 

Total marine shore and continental shelf to 200 meters 9,130 45.5 percent. 

(6) Continental slope and deep sea benthic below 200 meters 1280. 6.4 per- 
cent. Important components of this group are Macruridae, and species of Brotu- 
lidae, Zoarcidae, Apodes, and Scorpaeni formes. Contrary to the opinion of 
Greenwood et al. ( 1966), I do not believe that this group or group 8 contains a 
great number of unknown species. Fishes of these groups occupy a vast amount 
of space; however, conditions are relatively so stable and uniform that niches 
arc correspondingly few. 

(7) Epipelagic (high seas) above 200 meters 255. 1.3 percent. Important 



344 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

groups in this category are Scombroidei and Synentognathi. These fishes are 
mostly mobile, living in an environment that offers few niches. The small number 
of species is scarcely remarkable. 

(8) Deep pelagic below 200 meters (including mesopelagic and bathy- 
pelagic) 1010. 5.0 percent. Clupeiform and myctophiform fishes are the chief 
constituents of this category. Probably more space is occupied by this group 
than by any other, yet the number of species is small. The environment is poor 
in niches and in energy; it is surprising that the fauna is not smaller. 

SOME CONCLUSIONS 

The number of species in any one of the 8 categories seems to be chiefly re- 
lated to the degree of isolation possible. Certainly tropical reefs, great river 
deltas, and major river drainages have contributed a great variety of habitats 
and ecological niches which are reflected in the high percentage of species found 
in freshwater and along tropical shores. 

The most important regions economically (though not necessarily in terms of 
biological productivity) are the cooler water shelf areas and the epi pelagic, both 
regions with relatively few species. 

A final conclusion concerns the freshwater fishes. In view of the high per- 
centage of fishes found in freshwater and man's increasing modification of this 
environment throughout the world, it is vital that research be drastically increased 
on the basic systematics of freshwater fishes while this is still possible. 

ACKNOWLEDGMENTS 

It is a pleasure indeed to acknowledge the cooperation that I have received 
from my colleagues. Whatever value this paper may have is due to their contri- 
butions. I thank R. Bailey, P. Banarescu, R. Behnke, F. Berry, J. Bohlke, R. 
Bolin, M. Bradbury, J. Briggs, W. Burgess, D. Caldwell, B. Collette, E. Cross- 
man, W. Davis, H. DeWitt, W. Eschmeyer, J. Garrick, R. Gibbs, W. Gosline, D. 
Greenfield, P. H. Greenwood, M. Grey, R. Haedrich, E. Herald, L. Knapp, E. 
Lachner, R. Lavenberg, N. B. Marshall, H. McCully, R. McDowall, G. Mead, 
A. G. K. Menon, G. Miller, G. S. Myers, T. Nalbant, N. Parin, J. Randall, W. 
Richards, L. Rivas, C. R. Robins, R. Rofen, D. Rosen, R. Rosenblatt, L. Schultz, 
W. B. Scott, V. Springer, R. Suttkus, A. N. Svetovidov, W. R. Taylor, J. Tyler, 
E. Trewavas, B. Walker, V. Walters, A. Wheeler, N. Wilimovsky, and L. Woods. 

LITERATURE CITED 

Bailey, R. M. 

1960. Pisces (zoology), pp. 242-243 in McGraw-Hill Encyclopedia of Science and Tech- 
nology. Vol. 10. New York, McGraw-Hill. 
Greenwood, P. H. et al 

1966. Phyletic studies of teleostean fishes with a provisional classification of hving 



Vol. XXXVIII] COHEN: RECENT FISHES 345 

forms. Bulletin of the American Museum of Natural History, vol. 131, pp. 
339-456. 
Marshall, N. B. 

1965. The life of fishes. London, Weidenfeld and Nicolson. 402 pp. 
XORMAN, J. R. 

1963. \ histoiy of fishes. 2nd edition by P. H. Greenwood. New York, Hill and Wang, 
xxxi + 398 pp. 
Myers, G. S. 

1949. Salt-tolerance of fresh-water fish groups in relation to zoogeographical problems. 

Bijdragen tot de Dierkunde, vol. 28, pp. 315-322. 
1958. Trends in the evolution of teleostean fishes. Stanford Ichthyological Bulletin, vol. 
7, pp. 2 7-30. 

SCHULTZ, L. P. 

1965. Fishes and how they live, Chapt. 1 in Wondrous world of fishes. Washington 
D. C, National Geographic Society. 367 pp. 

SCHULTZ, L. P. AND E. M. SXERN 

1948. The ways of fishes. New York, Van Nostrand. xii + 264 pp. 



PROCEEDINGS 

OF THE 

CALIFORNIA ACADEMY OF SCIENCES 

FOURTH SERIES 



Festschrift for George Sprague Myers 



Vol. XXXVIII, No. 18, pp. 347-362; 8 figs.; 3 tables. December 31, 1970 



DESCRIPTION OF A NEW SUBSPECIES OF 

RHABDOPHIS AURICULATA IN THE 

PHILIPPINES, WITH COMMENTS ON THE 

ZOOGEOGRAPHY OF MINDANAO ISLAND 

By 

Alan E. Leviton 
California Academy of Sciences, San Francisco, California 94118 

During recent investigations on the snakes of the Philippine Islands, a study 
begun more than 15 years ago under the guidance of Professor George S. Myers, 
it became apparent that the island of Mindanao is inhabited by two clearly dis- 
tinguishable populations of the diminutive natricine snake, Rhabdophis auricu- 
lata. That the two populations of this species occupy different but contiguous 
parts of Mindanao make this discovery the more interesting, inasmuch as it 
helps shed further light on the paleogeography of the island. 

The type specimen of Rhabdophis auriciilata, a species described by GUnther 
in 1858, was collected by Hugh Cuming, its locality being given simply as 
"Philippine Islands." Miss Alice G. C. Grandison of the British Museum kindly 
examined GUnther's type, compared it with the figure published by Boulenger 
(1893, pi. 17, fig. 1) [fig. 1], prepared a sketch [fig. 2] of the color pattern at 
the angle of the jaw, and responded that Boulenger 's figure could well have been 
based on the type specimen. With Miss Grandison's sketch at hand, and after 
examining Boulenger's figure with care, I do not doubt that GUnther's type is 
based on a specimen drawn from the eastern Mindanao population. I have seen 
specimens from Davao Province that closely approximate the Giinther type, and 

[347] 



348 



CALIFORNIA ACADEMY OF SCIENCES [Proc. 4Tn Ser. 






•i.-x. 



Figure 1. Figure of Rhabdophis auriculata published by Boulenger (1893). 




Figure 2. Sketch of scale and color pattern at angle of jaw on left side of holotype. 



Vol. XXXVIII] LEVITON: RHABDOPHIS AURICULATA 349 

therefore, to insure stability of nomenclature, restrict the type locality of 
Tropidonotus auriculatus Gijnther to Mt. Apo, Davao Province, Mindanao Island. 
Specimens of Rhabdophis auriculata from Zamboanga del Norte, Zamboanga 
del Sur, Misamis Occidental, and Bukidnon provinces of western Mindanao 
[fig. 3] differ from animals from the Davao-Agusan region in color pattern. 
These animals are referred to a new taxon, named in honor of Professor Myers, 
who has for many years concerned himself with the zoogeography of the Philip- 
pines, especially Mindanao Island: 

Rhabdophis auriculata myersi Levi ton, new subspecies. 

(Figures 4-6.) 

Tropidonotus auriculatus, Boulenger, 1893, Cat. Snakes British Mus., vol. 1, p. 261 (part; 

Mindanao [Pasanaca, Zamboanga City]). 
Matrix auriculata, Taylor, 1922, Philippine Jour. Sci., vol. 21, p. 294 (Basilan [Port 

Holland]); 1923, Philippine Jour. Sci., vol. 22, p. 542 (Basilan, Mindanao [Zamboanga 

City]). 

Diagnosis. Maxillary teeth 27-32, last 3 or 4 enlarged but not separated 
by diastemata from others; head short, distinct from neck; scales in 17 longi- 
tudinal rows on anterior third body, reducing to 15 at about level of 80th ventral 
plate; usually 3 labials bordering orbit; ventrals 143-162; subcaudals 80-93; 
hemipenes extending to end of 6th subcaudal plate, forked at end of 4th-mid 5th 
plate; sulcus forked; proximal portion opposite sulcus with longitudinal plicae, 
portion bordering sulcus spinose; 2 large basal spines present; distal forks 
uniformly spinose; ventrolateral black stripe extending forward to suture of 
8th and 9th labials, continuing as diagonal stripe to corner of eye; lateral hght 
stripe extending uninterrupted (rarely interrupted at angle of jaw by black spot 
on 8th upper labial) along side of body to anterior temporals. Measurements 
(of largest male and largest female, in mm.): S s-v 342, tail 138; ? s-v 388, 
tail 165. 

HoLOTYPE. CAS-SU 23391, an adult male, taken near Masawan, approxi- 
mately 14 km. southeast of Buena Suerte, New Pifian, on west side of Dapitan 
Peak, Misamis Occidental Province, Mindanao Island, at an altitude of 4700 
feet (1433 m.) on 13 April 1959 by Dr. Angel Alcala (figs. 4-5). 

Paratypes (61). MINDANAO: Misamis Occidental Province: West 
side of Dapitan Peak: Masawan area, approximately 14 km. southeast of Buena 
Suerte, New Pirian, 26 March-20 April 1959 (CAS-SU 23372-23374, 23392- 
23394, 23397 [elev. 4400 ft. (1340 m.)]; 23375 [elev. 4400-4500 ft. (1340- 
1370 m.)]; 23376-23380, 23390 [elev. 4500 ft. (1370 m.)]; 23381-23382 
[elev. 4700 ft. (1430 m.)] ; 23383 [elev. 4300 ft. (1310 m.)]). Approximately 
2 km. east of Masawan and 15 km. southeast of Buena Suerte, New Pinan, 5 
April 1959 (CAS-SU 23384 [4700 ft. (130 m.)]). Approximately 12 km. east 
of Masawan, 6 April 1959 (CAS-SU 23387 [elev. 5500 ft. (1670 m.)]). Ap- 



3S0 



CALIFORNIA ACADEMY OF SCIENCES 



Proc. 4x11 Ser. 




Figure 3. Known distribution of RItabdophis auriciilata on Mindanao Island. 



proximately 2 km. southeast of Masawan and 16 km. southeast of Buena Suerte, 
New Pifian, 12 April 1959 (CAS-SU 23389 [elev. 4800 ft. (1460 m.)|). Ap- 
proximately 2-3 km. southeast of Masawan, 19 April 1959 (CAS-SU 23395- 
23396 [elev. 5000-5200 ft. (1520-1580 m.)]). Ridge 11-12 km. southeast of 
Buena Suerte, New Pinan, 22 April 1959 (CAS-SU 23398 [elev. 4000 ft. (1220 
m.)]). Bank of Dapitan River, 11-12 km. southeast of Buena Suerte, New 
Piiian, 22> April 1959 (CAS-SU 23399 [elev. 4000 ft. (1220 m.)]). Bank of 
Dapitan River, approximately 13 km. southeast of Buena Suerte, 25 April 1959 
(CAS-SU 23400 [elev. 4200 ft. (1280 m.)]). Zamboanga del Norte Prov- 
ince: West side of Dapitan Peak: Gumay area, approximately 6 km. south- 
east of Buena Suerte, New Pifian, 28-29 April 1959 and 4-8 May 1959 (CAS-SU 
23401, 23403-23404, 23419 |elev. 2500 ft. (760 m.)J, 23402 |elev. 2400 ft. 
(730 m.)], 23426-23427 [elev. 2300 ft. (700 m.)[). Dapitan River area, ap- 
proximately 1 km. southeast of Gumay and 7 km. southeast of Buena Suerte, 
New Pifian, 30 April-1 May 1959 (CAS-SU 23405-23413, 23415-23416 [elev. 
2300 ft. (700 m.)], 23414 [elev. 2400 ft. (730 m.)]). Gumay Creek, approxi- 
mately 6 km. southeast of Buena Suerte, 2-6 May 1959 (CAS-SU 23417, 23422 



Vol. XXXVIII] LEVITON: RHABDOPHIS AURICULATA 



351 




Figure 4. Holotype of Rhabdophis aunciilata inyersi Leviton (dorsal view). 



[elev. 2300 ft. (700 m.) j ). Bank of Dapitan River, approximately 2 km. south- 
east of Gumay and 8 km. southeast of Buena Suerte, New Piiian, 4 May 1959 
(CAS-SU 23418 [elev. 2500 ft. (760 m.)]). Approximately 7 km. southeast of 
Buena Suerte, New Pirian, 5 May 1959 (CAS-SU 23420-23421 [elev. 2700- 
2800 ft. (820-850 m.)]). Approximately 10 km. southeast of Buena Suerte, 
New Piiian, 6 May 1959 (CAS-SU 23423-23425 |elev. 3500 ft. (1070 m.)]). 
Gumay, approximately 5-6 km. southeast of Buena Suerte, New Pirian, 26 
March 1959 (CAS-SU 23371 [elev. 2500 ft. (760 m.) | ). Gumay Creek, approx- 
imately 0.5 km. southv.'est of Gumay and 6 km. southeast of Buena Suerte, New 
Pirian, 8 May 1959 (CAS-SU 23428 [elev. 2400 ft. (730 m.) ] ). Mt. Malindang: 
Masawan, April-May 1956 (CAS-SU 19362-19364 [elev. 3500-4500 ft. (1067- 
1372 m.)]). Between Masawan and Gandawan, April 1956 (CAS-SU 19535). 
MisAMis Occidental and Zamboanga del Norte Provinces Boundary: 
West side oj Dapitan Peak: Approximately 4 km. northwest of Masawan and 
10 km. southeast of Buena Suerte, New Pifian, 9-11 April 1959 (CAS-SU 
23385-23386 [elev. 3400 ft. (1040 m.)], 23388 [elev. 3500 ft. (1070 m.)]). 



352 



CALIFORNIA ACADEMY OF SCIENCES 



[Proc. 4th Ser. 




Figure S. Holotype of Rhabdophis auriculata myersi Leviton (ventral view). 



Additional material examined (30). BASILAN: Abung-Abung, 5-25 
October 1921 (CAS 60332-60339). Port Holland, 5-25 October 1921 (CAS 
60468). BOHOL: Cantaub Sitio, Sierra Bullones, 30 April 1955 (CAS-SU 
18895). MINDANAO: Bukidnon Province: Del Monte Plantation, 22 
August 1940 (CAS-SU 12396). Zamboanga del Norte Province: Catagan, 
15 May 1906 (USNM 37390). Katipunan, 30 km. up the Dicayo River (FMNH 
68905-68906). Miatan, Katipunan (FMNH 68916-68917). Zamboanga del 
Sur Province: San Ramon, 10 July 1929 (FMNH 14948). Zamboanga City, 
23 September-6 October 1920 (CAS 62023-62029). PHILIPPINE ISLANDS: 
(FMNH 68909, USNM 37415-37418). 

Description of holotype. Measurements (in mm.): snout-vent 318; taiP 
122; ventrals 157; subcaudals 82. Anal plate divided; 8 upper labials, 3-5 
bordering orbit; nasal shield divided below^ nostril only; loreal 1; preocular 1; 
postoculars 3; temporals 2 + 2,\ dorsal scales reduce 17 (-4f89/91] )15; 
hemipenes extend to end of 6th subcaudal plate, forked at middle of 5th plate; 
sulcus forked; forked portions spinose; half of proximal portion spinose on other 



Vol. XXXVIII] 



LEVITON: RHABDOPHIS AURICULATA 



353 



Table 1. Comparison of ventral counts for samples of Rhabdophis auriculata. 



Locality 






Sex 


Mean ± S.D. ± S.E. 


Range 


Number 


Mindanao 














Zamboanga 


del 


Norte^ 


$ 


154.9 ± 2.6 ± 0.5 


150-160 


29 








5 


154.6 ± 2.5 ± 0.4 


150-162 


33 


Zamboanga 


del 


Sur 


S 


152.2 ± 2.3 ± 1.0 


149-146 


5 








9 


1S3.S 


153-154 


2 


Bukidnon 






2 


148 


— 


1 


Davao 






$ 


150.9 ± 3.1 ± 0.6 


143-155 


28 








5 


152.7 ±3.5 ±0.7 


146-160 


28 


Agusan 






S 


ISO.S 


149-153 


4 








5 


1S0.S 


147-154 


4 


Basilan 






$ 


146.7 


143-150 


3 











152.2 ±3.1 ± 1.3 


147-157 


6 


Bohol 






$ 


148 


— 


1 


Luzon 






9 


145 


— 


1 



^ Based on a large sample from the Buena Suerta area, on Mt. Dapitan, along the boundary between 
INIisarais Occidental and Zamboanga del Norte provinces. 



side of sulcus; walls opposite sulcus with longitudinal plicae; 2 very large basal 
spines present on either side of sulcus. Color pattern as described in diagnosis; 
black ventrolateral stripe extending uninterrupted to meet lower postocular 
stripe; black midventral stripe originating on 24th ventral. 

Variation. There is scarcely any variation in head and body scutellation 
beyond that already reported for ventral and subcaudal counts. Only two of 



Table 2. Comparison of subcaudal counts for samples of Rhabdophis auriculata. 



Locality 






Sex 


Mean ± S.D. ± S.E. 


Range 


Number 


Mindanao 














Zamboanga 


del 


Norte^ 


$ 


86.3 ± 3.1 ± 0.6 


80-92 


25 








9 


86.3 ± 2.8 ±0.5 


82-93 


28 


Zamboanga 


del 


Sur 


$ 


85.4 ± 2.4 ± 1.1 


82-89 


5 








9 


86.5 


83-90 


2 


Davao 






$ 


79.7 ±3.9 ±0.8 


71-86 


22 








9 


80.5 ± 2.9 ± 0.6 


73-87 


26 


Agusan 






$ 


82.3 


81-85 


4 








9 


81.7 


73-87 


3 


Basilan 






$ 


83.7 


80-86 


3 








9 


83.4 ± 1.9 ±0.8 


80-86 


5 


Bohol 






$ 


84 


— 


1 


Luzon 






9 


78 


— 


1 



354 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 




Figure 6. Side view of head of CAS-SU 19362, Rhabdophis auriculata myersi (drawn by 
Marilyn Kramer) . 



87 specimens had 9 upper labials on both sides, one had 7 on both sides, one 
had 7 on one side, 8 on the other, and one had 9 and 8 labials. In most 
specimens there were 2 anterior temporals; however, in more than % of the 
sample the upper anterior temporal is divided by a vertical suture; 3 posterior 
temporals are usually present, only eight specimens have 2, and three have 2 on 
one side, 3 on the other. No variation was observed in numbers of preoculars, 
postoculars, or loreals. The nasal shield was carefully scrutinized under high 
magnification of a binocular stereoscopic microscope, and in no instance was I 
able to detect a suture in the shield above the nostril; a vertical suture was 
always present, extending from below the nostril to the lower border of the 
shield. Therefore, the nasal is only partially divided. 

The color pattern is remarkably stable, too. The dorsal ground color, in 
preserved specimens, is dark brown to black. There is a prominent middorsal 
light streak originating just behind the parietals and most distinct on the anterior 
fourth of the body. A lateral white stripe extends the length of the body along 
the outer edges of the ventrals to the neck, then angles forward onto the upper 
anterior temporals, ending just behind the postoculars [figure 6]. This stripe 
becomes narrower and less distinct posteriorly. It extends uninterrupted onto 
the side of the head, there being no squarish black spot on the neck behind the 
angle of the jaw as in the typical form [figure 7] connecting the dorsal ground 
color with the ventrolateral black stripe. This latter stripe, formed by black 
spots on the ventrals, extends forward, usually uninterrupted, onto the suture 
of the 8th and 9th lower labial and thence meets the lower postocular black bar 
which extends onto the 7th upper labial. Occasionally the stripe is interrupted 
for one scale width immediately behind the 8th lower labial. Otherwise it ex- 
tends posteriorly along the sides of the ventrals, broadening and usually coalesc- 
ing with the black middorsal stripe which originates on the 4th to 23rd ventral, 
extends posteriorly, becoming broad and joining the lateral black stripes. The 



Vol. XXXVIII] LEV1T0\: RHABDOPHIS AURICULATA 355 




Figure 7. Side view of head of CM 2592, Rhabdophis a. auriculata (drawn by Marilyn 
Kramer). 

venter is thus posteriorly almost entirely black except for a series of paramedial 
white spots, sometimes forming stripes. 

Ecological notes. The sample from Davao Province is large and is made 
up of both young and adult specimens. The smallest gravid female measured 
299 mm. snout-vent length; young with umbilical scars still evident, though al- 
most completely closed, measured up to 200 mm., although most animals above 
150 mm. in snout-vent length showed no visible signs of the scar. The young, 
defined here arbitrarily as those with some evidence of an umbilical scar or less 
than 200 mm. in snout-vent length, were collected mostly during the months of 
October and November 1946 by members of the Hoogstraal Expedition to 
Mindanao. The fact that young animals measuring from 104 to 200 mm. were 
collected during this period suggests that some egg laying takes place almost year 
round. However, 11 of the 19 young measured between 104 and 125 mm., 
suggesting further that there is probably a peak reproductive period during late 
spring or early summer. Gravid females were collected in June and July as well 
as during the October-November period previously mentioned. At least two speci- 
mens of R. a. myersi from northern Zamboanga, collected in April, were also 
gravid. Since most of the animals from the Zamboanga Peninsula were collected 
during April and early May, and those from Basilan in late October, this would 
add weight to my argument, based on the survey of the Davao sample, that a 
peak in egg laying takes place in late spring, possibly late May or June. Six of 
the sample measured less than 150 mm. snout-vent length, the smallest being a 
young female, 121 mm. from snout to vent. The rest of the sample divides up 
as follows: 150-199 m.m. snout-vent, 19 specimens; 200-249 mm., 8; 250-299 
mm., 13; 300-349 mm., 6; 350-399 mm.. 6; and 400 and over, 1. 

Only amphibian remains were found in the gut, mostly frogs, though in two 
cases there were tadpoles, and in three, masses of gelatinous frog eggs. In three 
specimens of R. a. auriculata from Davao Province, the frogs were identified as 
Oreophryne, a small terrestrial species found in southern Mindanao living in 



356 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

moss growing on logs or trees, under bark or in leaf axils (Inger, 1954, p. 447). 
Two specimens of R. a. myersi from near Buena Suerte had identifiable frog 
remains in the gut, one referable to Rana magna and the other to A nsonia niuel- 
leri. It is not too surprising that these frogs are eaten by R. aurkulata, consid- 
ering the diminutive size of the snake and of the frog, adults of which measure 
from 17.2-21.7 mm. (Inger, 1954, p. 447). 

The snakes for which we have adequate data are found at elevations from 
2800 to about 6400 feet (850-1950 m.) in the mountains of Davao Province 
(Mt. McKinley and Mt. Apo regions). In like manner, specimens of R. a. myersi 
from the northern part of the Zamboanga Peninsula were taken largely between 
2300 feet and 4500 feet (700-1370 m.). The type and those of the paratyiDic 
series for which data are available were taken from beneath rocks in dry river 
beds, while others were collected in brush or among other vegetation, usually 
near water courses, even if dry at that time of year (late September through 
November). A Basilan series of R. a. myersi collected by Taylor at Abung- 
Abung appear to come from near sea level. Also Taylor obtained a series from 
Zamboanga City, at or near sea level. However, the hills immediately behind 
Abung-Abung may in fact have been the source of these specimens. The 
Bunawan, Agusan Province, series collected by Taylor in 1921 was taken in the 
Agusan River valley, in the vicinity of water, usually from beneath leaves or logs 
at the edge of a small swamp (Taylor, 1922, p. 90). This locality lies below 
500 feet. Since Taylor's data seem quite reliable, the vertical distribution of 
R. auriculata is more than 600 feet, this being true for both nominal forms. 

Rhabdophis auriculata auriculata (Giinther). 

Tropidonohis auriculatiis Gunther, 1858, Cat. Col. Snakes British Mus., p. 80 (type locality: 
Philippine Ids. [restricted here to Mt. Apo, Davao Province, Mindanao] ; type in British 
Museum). Peters, 1861, Monatsb. Akad. wiss. Berlin, 1861, p. 687 (Samar). Muller, 
1883, Verh. Naturf. Ges. Basel, vol. 7, p. 286. Boettger, 1886, Ber. Senckenberg Naturf. 
Ges., p. 108; 1898, Kat. Rept. Samml. Senckenberg Naturf. Gesell., Schlangen, p. 28 
(Leyte). Fischer, 1885, Jahrb. Hamburg wiss. Anst., vol. 2, p. 80 (Siid-Mindanao). 
BotiLENGER, 1893, Cat. Snakes British Mus., vol. 1, p. 261, pi. 17, fig. 1 (part; "PhiHp- 
pines" [type specimen]). 

Natrix auriculata, Griffin, 1911, Philippine Jour. Sci., sec. D, vol. 6, p. 257 (part; distribu- 
tion compiled). Taylor, 1922, Snakes Philippine Ids., p. 89, pi. 4, figs. 2-4 (Mindanao 
[Bunawan, Agusan], Samar). 

Diagnosis. See Rhabdophis a. myersi except as follows: ventrals 143-160; 
subcaudals 71-87; ventrolateral black stripe extends forward to side of neck, 
does not extend onto throat, does not meet dark lower postocular diagonal 
stripe; light lateral stripe usually interrupted on side of neck behind angle of 
jaw by squarish black spot joining black ventrolateral stripe to dark dorsal 
color. IMeasurements (of largest male and largest female, in mm.): i s-v 352, 
tail 123; 5 s-v 372, tail 140. 



Vol. XXXVIII] LEVITON: RHABDOPHIS AURICULATA 357 

Table 3. Variation in head scutellation in Rhabdophis auricula ta auriculata. 



Preoculars 1-1=*(42) 2-2(3) 2-1(2) 1-2(3) 

Postoculars 3-3(41) 3-2(2) 4-3(1) 4-4(1) 3-4(1) 2-3(1) 2-2(3) 

Upper labials 8-8(46) 7-7(1) 

Temporals 2+3*(25) 2+2(12) 2+3/2 + 2(5) 2 + 2/2 + 3(4) 

1 + 3/2+3(1) 1 + 2(1) 2 + 1(1) 2+2+3(1) 

^ First number is count on right side, second number is for left side. 

* 2 + 3 indicates anterior + posterior temporals for both sides; 2 + 3/2 + 2 are counts for right and left 
sides. 

Range. Leyte-. Mindanao: Agusan Province (Bunawan); Cotabato Prov- 
ince (Burungkot) ; Davao Province (Mt. Apo [Baclayan, Todaya], Mt. McKin- 
ley). Samar'-. Luzonr. 

Material examined (86). LUZON: (CAS 15231). MINDANAO. 
Agusan Province: Bunawan (CM 2590-2598). Cotabato Province: 
Burungkot (FMNH 53345 felev. 1500 ft. (460 m.)]). Davao Province: 
Baganga River (USNM 34734-34735). Mt. Apo (FMNH 15016; 53322- 
53323 [elev. 2800 ft. (850 m.)]: 53326 [elev. 5500 ft. (1670 m.)]; USNM 
34713 [2 July 1904; elev. 4000 ft. (1220 m.)]; 34719 [3 July 1904; elev. 5000 
ft. (1520 m.)]; 34712, 34714-34718, 34720-34727, 34770-34771 [June-July 
1904; elev. 6000 ft. (1829 m.)]; 34728-34734). Todaya, Mt. Apo, November 
1946 (FMNH 53316-53321, 53324-53325, 53329-53344 [elev. 2800 ft. (850 
m.)]). Baclayan, Mt. Apo, 11 November 1946 (FMNH 53327-53328 [elev. 
6500 ft. (1980 m.)]). Mt. McKinley, 8 August-19 September 1946 (FMNH 
53299-53314 [elev. 300(^6400 ft. (910-1950 m.)]. 53315 [elev. 6300 ft. 
(1920 m.)], 53346 [elev. 3400 ft. (1040m.)]). 

Variation. Variation among the specimens examined is minimal. Ventral 
and subcaudal counts are summarized in tables 1 and 2. Variations in head 
scutellation are summarized in table 3. 

In color pattern in the typical form, variation is limited also. The Hght mid- 
dorsal stripe is present in 76 percent of the sample. However, its presence or ab- 
sence is not correlated with distribution, sex, or age group. The ventrolateral 
stripe is usually joined to the dorsal color by a vertical or almost vertical black 
bar on the neck behind the angle of the jaw in 83 percent of the sample. In 10 
percent the vertical bar does not contact the dorsal pattern, being separated 
by about one-half of a scale width. In only four animals — one from Burungkot, 
Cotabato Province, one from Mt. McKinley, and two from Mt. Apo, Davao Prov- 



» Specimens from Leyte and Samar were reported by Boettger (1898, p. 28) and Peters (1861, p. 687) 
respectively. The specimens on which these records were based are probably extant, Peters' at the Berlin 
Museum, Boettger 's at Senckenberg, and should be examined. I have not done so. 

The Luzon specimen is in the collection of the California Academy of Sciences, CAS 15231. Formerly 
it was in the Museo Santo Tomas and sent to the Academy around 1909 by Dr. J. C. Thompson, who said it 
probably came from Luzon. Indeed, the specimen may have originated on Luzon, but confirmation is needed. 
The specimen has an interrupted light lateral stripe and therefore is most similar to the eastern Mindanao 
population. 



358 CALIFORNIA ACADEMY OF SCIENCES LProc. 4th Ser. 

ince — is the bar absent. In most animals the ventrolateral stripe stops well short 
of the commissure of the mouth, being separated from the subpostocular blotch 
or stripe by at least 1 and usually 2 to 4 scales. 

Remarks. Rhabdophis auricidata is an unusually small natricine. I hesitate 
to venture an opinion on its relationships now, inasmuch as I haven't seen several 
Bornean forms such as Rhabdophis sarawacensis (Giinther) with which, based 
on available descriptions, it seems to agree. Although R. auriculata myersi is 
readily separable from R. a. auricidata on the basis of color pattern alone, an 
examination of data for ventral and subcaudal counts is revealing (table 1). A 
comparison of the Zamboanga-Misamis sample with the eastern Mindanao 
samples does not indicate a statistically significant level of difference in these 
counts, yet inspection of the means and ranges, especially of subcaudal shields, 
suggests the populations do indeed differ, though there is a substantial overlap. 
Perhaps most interesting is the fact that in subcaudal counts both Basilan and 
Bohol samples agree with the western Mindanao population, and though this is 
not confirmed by similar close agreement of ventral counts, even here the counts 
are well within the range of the western Mindanao sample. Of course, in color 
pattern both Basilan and Bohol samples are readily referrable to the typical 
subspecies. 

A further instructive comparison brings out the fact that the means and 
ranges of subcaudal counts for the Davao-Cotabato and Bunawan samples and 
the counts for the one specimen from Luzon are lower than for the typical form. 

Mindanao has long been of considerable interest to biogeographers concerned 
with the Philippine fauna and flora. Unquestionably it has served as a principal 
route for faunal and floral movements to the northern islands. Also it has had a 
complex post-Miocene history, which is still not completely unraveled. To the 
extent that we must deal with the matter here, it is sufficient to note that a 
substantial body of evidence, particularly in the form of coral reefs and numerous 
terraces, indicate clearly that, during the early and mid-Pleistocene, and possibly 
during the Pliocene, too, the island was divided into at least five and possibly 
seven smaller islands (fig. 8). These islands have been detailed by Dickerson 
(1928, pp. 85 87, fig. 19). Inger (1954, p. 453), in a paleographic map portray- 
ing the probable land areas during the lower Miocene based on Umbgrove (1938), 
suggests that even at that date Mindanao was probably not a single island, but 
rather a series of at least four island masses. The distribution of certain elements 
of the flora and fauna (see Merrill, 1928, p. 290; Cook, 1928, p. 269; Dickerson, 
1928, p. 295) suggests at least a division between the Zamboanga Peninsula and 
the rest of Mindanao. Further, these authors suggest a greater faunal and floral 
similarity between the Cotabato-Agusan-Surigao regions than between Cotabato 
and Zamboanga. This probably indicates that the Pleistocene islands depicted 
by Dickerson, i.e. Cotabato, Agusan, and Surigao, were connected by emerging 



Vol. XXXVIIIl 



LEVITON: RHABDOPHIS AURICULATA 



359 




Sarangani Bay 



Figure 8. "Some probable Pleistocene islands in Mindanao," from Dickerson (1928, 
p. 86, fig. 19). 



subaerial valleys before the Panguil-Illana Isthmus connecting the Zamboanga 
Peninsula with the Bukidnon-Lanao region emerged. 

In an earlier paper I pointed out that the populations of Cyclocorus nuchalis 
from eastern and western Mindanao differed significantly from one another 
(Leviton, 1967). This was a clue that suggested that in all probabiUty other 
species of snakes might show similar patterns which could be related to the 
Plio-Pleistocene paleogeographic history of Mindanao. A careful study of 
Rhabdophis aurkulata on Mindanao would seem to confirm this supposition. 
Populations of this species from Cotabato and Davao are most similar to the 
Surigao-Agusan populations I have seen and quite distinct from animals from 
the Zamboanga Peninsula. Samples of R. aurkulata from Bukidnon and Lanao 
provinces, on the other hand, are indistinguishable from those from the Zambo- 
anga Peninsula and Basilan Island, raising the possibility that western Mindanao, 



360 CALIFORNIA ACADEMY OF SCIENCES LProc. 4th Ser. 

composed of the "Pleistocene islands" of Lano and Zamboanga, were joined to 
one another and separated from the eastern complex of islands until fairly late in 
the Pleistocene by a persistent Macajalar-IUana Bay seaway. 

Indeed, it may have been the persistence of this seaway that prevented any 
significant eastward movement of the Lanao freshwater fishes. Significantly, only 
one species of Puntius is known from eastern Mindanao, P. binotatus, a Bornean 
species. Of course the seaway no longer exists and certainly there has been 
sufficient time for faunal movements throughout the island. Terrestrial animals 
can move with greater freedom than those that must wait for stream-capture in 
order to go from one drainage basin to another, so it is not surprising to find, 
for example, a paucity of species of cyprinid fishes in eastern Mindanao, though 
many forms are known from Lake Lanao and several from the Zamboanga Pen- 
insula. On the other hand, many subspecies of frogs are widely distributed 
throughout Mindanao, Samar, and Leyte. This distribution is obviously a more 
recent development, probably a late Pleistocene phenomenon. Indeed, it is be- 
coming more apparent that several faunal movements and radiations can be 
identified in Mindanao and are correlated with its post-Oligocene geological 
history. 

Quite clearly, a more extensive and careful study of populations of animals 
and plants on Mindanao, especially the montane forms, should give us consider- 
able insight into the late geological history of that region. 

ACKNOWLEDGMENTS 

The writer wishes to express his appreciation for the loan of specimens to Mr. 
Neil B. Richmond and Dr. Clarence J. McCoy, Carnegie Museum; Dr. Robert 
F. Inger and Mr. Hymen Marx, Field Museum of Natural History; and Dr. 
James A. Peters, U.S. National Museum. 

The following abbreviations are used : CAS — California Academy of Sciences ; 
CAS-SU — California Academy of Sciences-Stanford University (for specimens 
formerly housed at Stanford University and registered in the Stanford cata- 
logues) ; CM — Carnegie Museum; FMNH — Field Museum of Natural History: 
USNM— United States National Museum. 

LITERATURE CITED 

BOETTGER, OSKAR 

1898. Katalog der Reptilien Sammlung in Museum der Senckenbergischen Naturfor- 
schenden Gesellschaft in Frankfurt am Main. 2. Teil. Schlangen. Frankfurt-am- 
Main, ix + 160 pp. 

BOULENGER, GeORGE A. 

1893. Catalogue of the snakes in the British Museum (Natural History). Volume 1. 

London, xiii -|- 448 pp., 28 pis. 
Cooke, K. H. 

1928. Land mollusks of the Philippines. In: Dickerson, Roy (ed.). Distribution of Hfe 

in the Philippines. Monograph of the Bureau of Science, Manila, Philippines, 

no. 21, pp. 267-272. 



Vol. XXXVIII] LEVITON : RHABDOPHIS AURICULATA 361 

DiCKERSON, Roy 

1928. Tertiary and Quarternary paleogeography of the Philippines. Monograph of the 
Bureau of Science, Manila, Phihppines, no. 21, pp. 76-96. 
GuNTHER, Albert 

1858. Catalogue of colubrine snakes in the collection of the British Museum. London, 
xvi + 281 pp. 
Inger, Robert F. 

1954. Systematics and zoogeography of Philippine Amphibia. Fieldiana: Zoology, vol. 
33, pp. 181-531. 
Leviton, Alan E. 

1967. Contribution to a review of Philippine snakes, IX. The snakes of the genus 
Cyclocorns. Philippine Journal of Science, vol. 94, pp. 519-533. 
Merrill, Elmer D. 

1928. Floral provinces and subprovinces of the Philippines. In: Dickerson, Roy (ed.). 
Review of the origins of the biologic provinces of the Phihppines. Monograph 
of the Bureau of Science, Manila, Philippines, no. 21, pp. 289-294. 
Peters, Wilhelm C. 

1861. Eine zweite Ubersicht (vergl. Monatsberichte 1859, s. 269) der von Hrn. F. Jagor 
auf Malacca, Java, Borneo und den Philippinen gesammelten und dem Kgl. 
zoologischen Museum iiberstandten Schlangen. Monatsberichte der Koniglich 
preussischen der Akademie der wissenschaften zu Berlin, 1861, pp. 683-691. 
Umbgrove, H. J. F. 

1938. Geologic history of the East Indies. Bulletin of the American .Association of 
Petroleum Geologists, vol. 22, pp. 1-70, 1 pi. 



PROCEEDINGS 

OF THE 

CALIFORNIA ACADEMY OF SCIENCES 

FOURTH SERIES 



Festschrift for George Sprague Myers 



Vol. XXXVIII, No. 19, pp. 363-382; 7 figs. December 31, 1970 



A REINTERPRETATION OF THE 

TELEOSTEAN FISH ORDER 

GOBIESOCIFORMES 

By 

William A. Gosline 
University of Haivaii, Honohiht 

An intensive effort to interpret relationships among the old group "Jugulares" 
(Linnaeus, 1758, p. 249; Jordan, 1923, p. 228; etc.) led to a consideration of the 
Callionymidae and Draconettidae. For reasons dealt with below, the conclusion 
was reached that these two families (I do not agree with Davis, 1966, that they 
should be combined) are specialized derivatives of the notothenioid section of the 
perciform suborder Blennioidei (Gosline, 1968). Since, however, the Draconet- 
tidae and Callionymidae are morphologically well differentiated from the noto- 
thenioids, it appears best to remove them from the Perciformes entirely. Inves- 
tigation also suggested that the Gobiesocidae has evolved from the notothenioid 
section of the perciform suborder Blennioidei and in small part at least over the 
same route as the draconettids and callionymids. The Callionymidae, Draconet- 
tidae, and Gobiesocidae are therefore combined here in the order Gobiesociformes. 

The systematic position of the Callionymidae and Draconettidae has never 
been the subject of direct investigation. \'arious views concerning the relation- 
ships of these two families have, however, been suggested. Boulenger (1904, p. 
708) included both the Callionymidae and Gobiesocidae in his Division Jugu- 
lares, and under his account of the Gobiesocidae stated: "The position of the 
ventral fins suggests, at first glance, affinity with the Callionymidae, and a com- 

[363] 



364 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

parison of the skeletons of these two types has convinced me that they are really 
related to each other, though highly modified in different directions." (My own 
conclusions are essentially those of Boulenger.) Starks (1905, p. 302) in con- 
nection with his account of the gobiescoid Caularchus | =Gobiesox] inacandricus 
wrote: "The Callionymidae, however, possess some important characters not 
possessed by the Gobiesocidae, and these probably more than counterbalance the 
characters held in common." Regan in 1913 (pp. 144, 145) placed the Calliony- 
midae and Draconettidae in the "Division Callionymiformes" of the perciform 
suborder Percoidei. He stated that the Callionymidae "may be related to the 
Pinguipedidae, but is much more specialized [a suggestion with which I also 
agree]. The Gobiesocidae differ in many characters of importance." Referring 
again to the Callionymidae, Starks (1923, p. 267) said: "The osteology shows, 
however, that this family on account of several rather extraordinary and unique 
characters should be segregated in a suborder coordinate in value with the Batra- 
choid fishes." Regan in 1929 also recognized the Callionymoidei as a perciform 
suborder. The most recent comment on the systematic position of the Draconet- 
tidae and Callionymidae is that of Briggs and Berry (1959, p. 125). They sum- 
marized as follows: "Considering the paucity of our knowledge about these two 
families and their relationships with other percomorph groups, we see no present 
need for setting them aside in a separate suborder. Their morphology is no more 
peculiar than that of several other families that are traditionally retained without 
subordinal recognition within the vast assemblage of the Percomorphi." 

The best and most complete account of the anatomy of the Gobiesocidae re- 
mains that of Guitel (1889). However, Guitel draws no conclusions regarding 
gobiesocid relationships within the Acanthopterygii. Since the days of Starks 
(1905)^ and Regan ( 1909) the family has generally been allocated to an order of 
its own. In his monograph of the family, Briggs (1955, p. 7) wrote: "The Xeno- 
pterygii [=Gobiesociformes] seems to be most closely allied to the Haplodoci 
(batrachoids) but there is also some resemblance to the Callionymoidea. The 
order may be considered a highly specialized derivative of some still unknown 
primitive percomorph stock." McAllister (1968, p. 165) also suggests a gobie- 
socid-batrachoidid relationship. Apparently on the assumption that such exists 
Greenwood et al. (1966, pp. 389, 397) have assigned the Gobiesociformes to the 
superorder Paracanthopterygii, thus separating the group superordinally from 
the callionymoids. 

Under the circumstances, it first seems advisable to discuss the possible rela- 
tionship between the gobiesocoid and batrachoid fishes. Though both groups 
hold certain characteristics in common, e.g., the usually scaleless body, the flat- 
tened head, anterior pelvics, incomplete circumorbital series, etc., it is my pro- 



1 Starks (1905, p. 292) attributed the creation of ordinal status for the gobiesocids to Gill, but neither 
Briggs (1955, p. 7) nor I have been able to find where Gill recognized more than subordinal rank for this 
family. 



Vol. XXXVIII] GOSLIXE: GOBIESOCIFORMES 365 

visional view that these similarities are the result of convergence. The batra- 
choid fishes differ from the Gobiesociformes, i.e., Callionymidae, Draconettidae, 
and Gobiesocidae, in the following features: 

In the batrachoid fishes the pelvic fins are fairly close together, small, and 
with 2 or 3 soft rays that are usually held out at an angle from the abdominal 
surface; in the Gobiesociformes the pelvic fins are wide apart, well developed 
(though highly specialized in the Gobiesocidae), and of 4 or 5 soft rays that are 
normally held flat against the body surface. In the batrachoids the upper hypu- 
rals have a peculiar intervertebral-like basal articulation with the rest of the 
caudal skeleton (Regan, 1912, fig. 2B) ; in the Gobiesociformes there is no such 
articulation. In the batrachoids the ascending process of the premaxillary has a 
movable basal articulation with the toothed portion, and a separate articular 
process of the premaxillary is well developed (Monod, 1960, fig. 49); in the 
Gobiesociformes the ascending and articular processes of the premaxillary have 
merged or fused and are firmly attached to the toothed portion. In the batra- 
choids there is no median ethmoid ossification; in the gobiesociform fishes a 
median ethmoid ossification is always present. Finally, the batrachoids have a 
peculiar gas bladder (S0rensen, 1884); in the Gobiesociformes there is no gas 
bladder. 

With regard to the postulated derivation of the Gobiesociformes from the 
superfamily Notothenioidae (containing the parapercids [ = mugiloidids], chei- 
marrichthyids, trichonotids, nototheniids, etc., see Gosline, 1968) of the perci- 
form suborder Blennioidei, the gobiesociform fishes have almost all of the diag- 
nostic notothenioid characteristics despite their high degree of specialization 
along other lines. 

Thus in the Gobiesociformes the head is always more or less flattened, some- 
times greatly so. The circumorbital ring of bones is incomplete. The medial 
tabulars are apparently lacking. There is a basisphenoid bone in Draconetta but 
not in the Callionymidae and Gobiesocidae. Flanges from the parasphenoid do 
not extend up in front of the prootics excluding the prootics from the internal 
cranial border of the orbit. (When, as in some gobiesocids and callionymids, the 
parasphenoid does have an upward expansion, this extends up between the mid- 
dle portion of the orbits, not in the form of a postorbital bar such as occurs for 
example, in the zoarcioid blennioids.) The pelvic fins are as noted above. (The 
Gobiesocidae, in which the pelvic fins form the anterior portion of the sucking 
disc, is the only group known to me in which such a disc extends well forward of 
the pectoral bases.) The pectoral actinosts are three or four in number. (The 
3 broad plate-like actinosts of the Callionymidae are closely duplicated in such 
notothenioid families as the Nototheniidae.) The dorsal and anal rays are equal 
in number to the vertebrae between them. The caudal fin is rounded or brush- 
like, with fewer than 1 5 branched rays. 

Additional notothenioid resemblances of the Gobiesociformes are as follows. 



366 



CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 







Vol. XXXVIII] GOSLINE: GOBIESOCIFORMES 367 

The ventral sucking disc of the Gobiesocidae would seem to be to some extent 
foreshadowed in the ridges on the flat ventral surface of the notothenioid family 
Cheimarrichthyidae. The notothenioid genera Prolatilus and Mugiloides are the 
only members of the suborder Blennioidei known to me with the draconettid 
supraoccipital crest and with the body musculature extending well forward over 
the top of the cranium. Seven branchiostegal rays, said to be present in some 
members of the Gobiesocidae (Briggs, 1955) also occur in a number of noto- 
thenioids, but rarely elsewhere among the Blennioidei. Finally, the opercular 
pecuUarities of Draconetta (fig. lA) are largely dupUcated in the notothenioid 
Harpagijer (fig. IB) and would seem to be foreshadowed in the more generaUzed 
notothenioid Pcrc/'erm (fig. IC). 

The anatomical account of the draconettids, callionymids, and gobiesocids 
which follows is based primarily on alizarin-stained and dissected specimens from 
the following lots : 

Callionymidae: Callionynms jlagris, 125 mm. in standard length (U. S. Na- 
tional Museum no. 71082); C. decoratus, 50 mm. (University of Hawaii no. 
2073) ; and Pogonymus pogognathus, 24 mm., paratype (UH 1626). 

Draconettidae: Draconetta acanthopoma, 75 mm. (USNM 156956). 

Gobiesocidae: Gobiesox nigripinnis , 70 mm. (USNM 131163), and Tra- 
chelochismus pinnulatus, 55 mm., an exchange specimen from New Zealand in 
the UH collections. 

The external features of various other species of Callionymidae and Gobie- 
socidae in the U. S. National Museum were examined during tenure of a Smith- 
sonian Research Associateship. I wish to express my deep obligation to the 
members of the Fish Division of that institution for help and facilities during 
that time and for sending me on loan the specimens of Draconetta acanthopoma 
listed above. 

General features. The head and body of callionymids, draconettids, and 
gobiesocids are always scaleless, although Ochiai (1963, p. 66) finds "degenerate 
scales" partly surrounding the lateral line canal of the callionymid Diplogrammus 
goramensis. In calHonymids the gill opening is a small hole ; in Draconetta it is 
larger, but the gill membranes are broadly attached to the isthmus; and in the 
Gobiesocidae the gill membranes may be attached to or free from the isthmus 
(Briggs, 1955). The widely separate pelvic fin bases are entirely in front of the 
broad pectoral bases, which extend far down the sides; in some callionymids and 



Figure 1. Right suspensorium and opercular bones, external view, of A, Draconetta 
acanthopoma; B, Harpagijer bispinis; C, Parapercis cephaloptinctata; D, Callionymns flagris; 
and E, Gobiesox nigripinnis. ec, Ectopterygoid ; hy, hyomandibular ; io, interopercle ; ms, 
mesopterygoid ; mt, metapterygoid ; op, opercle; pa, palatine; po, preopercle; and sy, sym- 
plectic. 




368 



CALIFORNIA ACADEMY OF SCIENCES 



[Proc. 4th Ser. 




Figure 2. CalUonyiiius flagris. Sketch of right side of head to show lateral line canals 
(dashed hnes) that are not enclosed in head bones. 



gobiesocids there is indeed a membrane extending from the innermost pelvic rays 
onto the outer surface of the pectoral fin. There is a short spinous dorsal in the 
Draconettidae and usually in the Callionymidae, but never in the Gobiesocidae. 

Fin structure. The Gobiesociformes show a transitional series from the 
usual percoid condition with spines and branched rays to that of the Gobiesocidae 
where the only spinous element is the flat outer pelvic plate and all the soft rays 
are simple. The loss of the spinous dorsal in this series has already been noted. 
As for soft rays, in the callionymid genera Yerutia and Synchiropus all of the 
soft dorsal rays may be branched (Schultz, 1960, p. 399) and in large specimens 
of Draconctta acanthoponia all of the anal rays are branched, but elsewhere the 
dorsal and anal rays are mostly or all simple. In Draconetta and the callionymids 
examined, most of the pectoral and the 5 pelvic rays are branched. I count 8 
branched caudal rays in Draconetta acanthoponia, 6 in Callionymus flagris. 

The lateral line system. Those portions of the lateralis system enclosed 
in head bones will be dealt with below. Here, only the peculiar extension of the 
lateralis system in the Callionymidae will be mentioned. Such extensions occur 
on both the head (fig. 2) and body. In Callionymus, the system includes such 
peculiar features as a commissure across the top of the caudal peduncle. On the 
head of the same genus the canals behind the frontals all lie superficial to the 
skull bones, extending across the surface of the pterotic and forming a complete 
supratemporal commissure that is not contained in extrascapulars. Again the 
preopercular canal, instead of running up within that bone, exits from its lower 
limb, passes out superficially across the preopercular spine, and then up over the 
flesh behind the preopercle (fig. 2). None of the peculiarities mentioned are 
found in either the Draconettidae or the Gobiesocidae, although in the Draconet- 
tidae there are membranous extensions of the lateralis system. 



Vol. XXXVIII] GOSLINE: GOBIESOCIFORMES 369 

Nasal apparatus. The nasal apparatus differs considerably among the 
gobiesociform fishes examined. It is most percoid-like in Gobiesox nigripinnis 
which has 2 nostrils, the anterior with a fringed flap and the posterior in a raised 
collar; these 2 nostrils lead into a nasal cavity, bordered mesially above by the 
nasal bone; the cavity contains a roundish nasal rosette. The nasal apparatus of 
Callionymus jlagris is about the same except that there is only 1 nostril on each 
side. In Draconetta acanthopoma there are 2 tubular nostrils but no nasal bone; 
the nostrils lead into the two ends of a flattened, hollow, fleshy pad which seems 
to contain no speciaHzed olfactory folds or lobes. 

The circumorbital bones. The circumorbital series in the Gobiesociformes 
is always reduced to the lacrimal bone. Behind the eye in Callionymus and Dra- 
conetta a membrane-enclosed canal exits from the main lateralis canal and extends 
downward. In Draconetta this canal is short, ending behind the eye; in Callio- 
nymus jlagris it extends forward below the eye towards the base of the lacrimal 
bone (fig. 2) but fails to connect with the lacrimal-enclosed canal. 

Jaw structure. The peculiarity of the upper jaw protrusion of Callionymus 
has been described by van Dobben ( 1935, pp. 47, 48) and by Kayser ( 1962 ) . In 
most percoids, the maxillary heads twist on their axes extruding the premaxillary 
articular processes before them like a squeezed cake of soap (van Dobben, 1935, 
pp. 10-13). In Callionymus the maxillary heads and associated cartilages and 
ligaments of the two sides form a ring around the long ascending processes of the 
premaxillaries. The ascending processes of the premaxillaries are free to move in 
and out within this ring. Upper jaw protrusion is entirely produced by the lower- 
ing of the mandible with the associated downward movement of the lateral end of 
the maxillary. Anatomically Callionymus is peculiar in having no articular proc- 
esses on the premaxillaries lateral to their ascending processes. 

The gobiesocids also have premaxillaries without articular processes (Guitel, 
1889, pi. 25, fig. 16, and Briggs, 1955, figs. 74-81). In Draconetta there are 
long, narrow, articular processes that are all but fused to the ascending processes. 
So far as I determine from preserved specimens, Draconetta and most gobiesocids 
use the same peculiar method of upper jaw protrusion that Callionymus does. In 
at least the gobiesocid genus Tomicodon, however, the upper jaw does not appear 
to be protrusile. 

Gill covers and suspensoria. With the extreme flattening of the head 
region that has taken place in the Callionymidae, Draconettidae, and Gobiesoci- 
dae, the operculum becomes squashed, so to speak, into a horizontally elongate 
structure. In at least some members of all three families, backwardly projecting 
spines are developed, but they are formed in different ways. 

As already noted, the opercular apparatus of Draconetta (fig. lA), with spi- 
nous opercles and subopercles, is essentially similar to that of the notothenioid 
Harpagijer (fig. IB), although in Harpagifer an additional support for the oper- 
cle has been added by extending a vertical strut up to an abutment against the 



370 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

cranium. How the spinous arrangement in Harpagifer and Draconetta might 
have originated is suggested by the basal notothenioid Parapercis (fig. IC). In 
Parapercis the opercle ends in the not unusual point; the subopercle has two 
structurally different sections, an upper, flap-like ossified membrane and a lower 
rigid plate ending posteriorly in a few serrations. Disappearance of the upper 
membranous portion of the subopercle and development of the lower would provide 
essentially the configuration of gill cover spines found in Harpagifer and Draco- 
netta. 

Now, if instead of developing the lower portion of the subopercle of Para- 
percis, the upper flap-like portion were enlarged, the lower eliminated, and the 
preopercle developed backward as a strong spine, the configuration found in Cal- 
lionymus (fig. ID) would result. 

To arrive at the gobiesocid-type opercle (fig. IE), one could hypothesize a 
form of Callionymus in which the subopercle loses its association with the inter- 
opercle and swings back onto the end of the opercle where it may form a spine in 
gobiesocids. 

The changes in opercular structure just described are reflected in the inter- 
opercle. This bone, fairly long in Parapercis and longer in Draconetta, is pulled 
out into a long weakly ossified tendon in Callionymus. In Gobiesox the inter- 
opercle is wholly concealed by the preopercle and does not reach the subopercle 
at all but terminates in an abutment against the rear of the hyoid apparatus, as 
in the Blenniidae; the interopercle is, however, better developed in the more 
primitive Trachelochismus , where it nearly reaches the subopercle. 

A last minor point about the gill cover structure of the Gobiesociformes 
should perhaps be made. In all three families those edges that are not rigid tend 
to have long, flexible bony fimbriae. 

The "squashing" of the opercle would also seem to have had an effect on the 
suspensoria of callionymids, draconettids, and gobiesocids. The preopercles of the 
callionymids (fig. ID) and gobiesocids (fig. IE) have been extruded backward, 
so to speak, and the hyomandibular, preopercle, and quadrate have come to form 
the three points of a triangle. Draconetta (fig. lA), however, has retained the 
usual configuration with the hyomandibular, preopercle, and quadrate all more 
or less in line. There is however no separate metapterygoid in the Draconettidae, 
Callionymidae, or Gobiesocidae. 

The connection between the palatine and the posterior portion of the suspen- 
sorium has become rather tenuous. In Draconetta (fig. lA) the palatine is at- 
tached to the quadrate by a long narrow strut composed of the ectopterygoid and 
mesopterygoid. In Callionymus (fig. ID) these last two bones seem to have fused, 
but the strut is still present. In Gobiesox (fig. IE) the palatine is only loosely 
connected with the rest of the suspensorium, the mesopterygoid is gone, and the 
minute ectopterygoid is only ligamentously attached to the palatine. 

The hyoid apparatus and gill arches. The hyoid apparatus is close to, 



Vol. XXXVIII] 



GOSLINE : GOBIESOCIFORMES 



371 




Figure 3. Hyoid and gill arches (1-5) in Gobiesox nigripinnis: A, hyoid arch and lower 
portions of the gill arches of the right side, from above; and B, the upper portions of the gill 
arches of the left side, from above, cb, Ceratobranchial; ch, ceratohyal; eb, epibranchial; eh, 
epihyal; gh, glossohyal; hb, hypobranchial ; hh, hypohyal; ph, upper pharyngeal tooth plate. 



and firmly connected by the anterior basibranchial with, the gill arches in Callto- 
nymus: in Draconetta and Gobiesox the hyoid arch is well separated from and 
unconnected with the other gill arches. In Callionymus and Draconetta, a well 
developed glossohyal extends forward from the hypohyals; in Gobiesox (fig. 3) 
the glossohyal is a small sliver of bone completely contained in the interspace 
between the hypohyals of the two sides. In Draconetta and Callionymus there 
are 6 branchiostegal rays on each side, in the gobiesocids 5-7 (Briggs, 1955, p. 9). 
In Draconetta, Callionymus, and Gobiesox there are 2 anterior branchiostegals 
attached to the inner surface of the hyoid arch; the other 4 close to its lower rim. 
In Draconetta and Callionymus the first 2 are short; in Gobiesox the first 3. In 
Draconetta 4 of the 6 branchiostegals are crowded back on the epihyal, in Callio- 
nymus 3, and in Gobiesox only 1 branchiostegal articulates with the epihyal. 

Among gobiesociform families the first spicular pharyngobranchial seems to 
have completely disappeared and there are never more than 2 sets of pharyngeal 
teeth on either side above. In Draconetta and Callionymus epihyal 2 extends up 
to the relatively small and narrow anterior tooth patch, while epihyals 3 and 4 
articulate with the broader, posterior pharyngeal tooth patch-; in these two genera 
epihyals 3 and 4 are closely but movably attached to one another. In Gobiesox 



- Staxks (1905. p. 302) stated that Callionymus had ''three superior pharyngeals on each side'' but in 1923 
(p. 269) he describes 2 upper pharyngeals of the same shape as noted here. 



372 



CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 





fhsc 



Figure 4. Cranium plus upper portion of pectoral girdle, right side, from above, of A, 
Gobiesox nigripinnis; B, Draconetta acanthopoma (only the upper surface of the rostral 
region is shown) ; and C, Callionymus flagris. Lateral line canals passing through cranial 
bones indicated by dashed lines, ca, Cartilage ; cl, cleithrum ; cv, cavity into which the ascend- 
ing processes of the premaxillaries extend ; et, lateral extrascapular ; ex, exoccipital ; fr, frontal ; 
le, lateral ethmoid; me, mesethmoid; pa, parietal; po, posttemporal ; pt, pterotic; su, supra- 
cleithrum; so, supraoccipital ; st, sphenotic; su, supraoccipital crest; and vo, vomer. 



(fig. 3) there are no chondrified or ossified basibranchials, and the gill arches are 
not interconnected below. Above, there is only 1 small pharyngeal tooth plate on 
each side; epihyals 2, 3 and 4 articulate with it, and epihyals 3 and 4 are rigidly 
united to one another. 

Skull. In Gobiesox (which Uves under rocks in the tidal zone) the head is 
broad with small eyes in strong laterally placed bony sockets. In Callionymus 
and Draconetta the eyes are close together on the top of the head. These differ- 
ences are strongly reflected in the crania. 

Lateral line and associated skull bones. In the Gobiesocidae the forward por- 
tion of the supraorbital canal on each side commences near the snout rim and 
passes back through the paired nasal and frontal bones. Between the wide-set 
eyes there is a complete, bone-enclosed frontal commissure (fig. 4A). In the Dra- 
conettidae and Callionymidae the narrow interorbital region has doubtless caused 
the fusion of the 2 supraorbital canals into a single median canal between the eyes 
(fig. 4B, C). Furthermore, in Draconetta the frontals themselves have fused into 
a single median bone. However, in the two species of Callionymus examined the 
frontals appear to be only partially fused, and in the callionymid Pogonymus, 



Vol. XXXVIII] GOSLINE: GOBIESOCIFORMES 373 

which has a somewhat broader interorbital area, I beheve I can see a suture com- 
pletely dividing the frontals. Anteriorly, the supraorbital canals of callionymids 
begin in the separate nasals as usual, but in Draconetta acanthopoma there are 
no nasal bones and the anterior median pore of the frontal canal is the anterior- 
most point in the supraorbital system. (In Draconetta oregona Da\ds, 1966, fig. 
2, shows the supraorbital canals as separating ahead of the eyes and extending 
forward on each side to just behind the nostril. Perhaps these anterior extensions 
of the supraorbital system in D. oregona are represented by fine ridges of flesh 
running over the same areas in D. acanthopoma.) 

Behind the frontals the lateral line canals of Callionymus lie superficial to the 
skull bones, as previously noted (fig. 1). In Gobiesox and Draconetta the tem- 
poral canals pass backward from the frontals through what appears to be the 
sphenotic and pterotic (fig. 4A, B). Passage of the lateral line through the pterotic 
is normal in fishes, but a canal in the sphenotic is not. Possibly the "sphenotic" 
canal of Gobiesox and Draconetta extends through a dermosphenotic which has 
become fused to the sphenotic. In Draconetta the lateral line canal passes back 
from the pterotic into a lateral tabular, where it gives off the membranous, incom- 
plete supratemporal commissure, and then into the posttemporal, where it ends. 
In Gobiesox the lateral line canal ends in the pterotic ; there is no tabular bone or 
posttemporal commissure. 

Ethmoid region of the skull. The peculiarities of the ethmoid region of the 
cranium of Callionymus (Starks, 1923, pp. 267-268) and of Draconetta can, I 
think, have developed through a pinching together of the broader, more normal 
ethmoid area of the Gobiesocidae. In the Gobiesocidae the ethmoid overlaps the 
vomer in the usual percoid fashion but lies behind the level of the lateral ethmoids 
(Guitel, 1889, pi. 25, fig. 1). In the narrower-snouted draconettids and callio- 
nymids the mesethmoid is completely separated from the vomer by cartilage and 
by the medial bases of the two lateral ethmoids which meet (fig. 4C) or nearly 
meet on the midline. In Draconetta (fig. 5) the mesethmoid is above and behind 
the lateral ethmoid bases, but in the callionymids it is entirely behind them. In 
both families the mesethmoid forms part of the orbital border. In the calHonymid 
Pogonymiis the ascending processes of the premaxillaries extend up and back over 
the rostral surface as usual; here the mesethmoid does not extend down into the 
interorbital space. But in Callionymus the ascending processes are more horizon- 
tal and their tips extend back into a medial rostral cavity; here the mesethmoid 
has been pushed down and back, as it were, into the infraorbital space (Starks, 
1923, pi. 4, fig. 5) . The same sort of thing seems to have happened in the chaeto- 
dontid percoids, as Starks (1926, p. 301, footnote 35) has noted. 

Upper surface of rear of skull. ^Nlajor differences on the upper surface of the 
skull posteriorly have to do with the extent to which it is covered by the body 
musculature. In Callionymus the rear face of the skull drops away abruptly, and 
no musculature at all extends forward ov^er its upper surface. The supraoccipital 



374 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 




Figure 5. Draconetta acanthopoma. Lateral view of forward end of cranium with only 
the base of the lateral ethmoid indicated. Cartilage stippled, fr, Frontal; io, infraorbital 
fenestra ; le, lateral ethmoid ; me, mesethmoid ; ra, parasphenoid ; and vo, vomer. 



extends back from the skull as a flat superficial cap (fig. 4C) the bottom of v^^hich 
forms a surface for muscular attachment. In Gobiesox the rear face of the skull 
slopes more obliquely and two large lateral lobes of musculature extend forward 
nearly to the rear borders of the eyes. The musculature does not extend forward 
over the central portion of the skull and there is no median crest. The flat supra- 
occipital bone (fig. 4A) in Gobiesox is pinched off into two parts by the overlap- 
ing parietals, but this is not true of at least certain other gobiesocid genera (see 
Guitel, 1889, pi. 25, fig. 1, and Starks, 1905, p. 283 — Starks's Caularchns equals 
Gobiesox and his Gobiesox equals Sicyases according to Briggs, 1955). In Dra- 
conetta (fig. 4B) the musculature extends forward along either side of the midline 
to just behind the eye, and a median crest extends forward on the supraoccipital 
and even a short way on the fused frontals. 

Sphenoid region of the skull. In Draconetta a pleurosphenoid and small 
basisphenoid bone are present; the two bones are, however, widely separated, the 
basisphenoid ending posteriorly in the membrane lining the orbits posteriorly. 
In neither Callionymus nor Gobiesox are pleurosphenoids or basisphenoids present. 

As is true of all notothenioids, there is little upward extension of the para- 
sphenoid into a postorbital bar, and the prootic borders the orbit in all the 
Gobiesociformes. In Gobiesox, however, the parasphenoid is considerably ex- 
panded anteriorly, forming a broad shelf below and between the orbits; this ex- 
pansion is greater than that of the parasphenoid just behind the orbits. 

Otic and occipital regions of the skull. In none of the Gobiesociformes 
examined is there an expanded auditory bulla. In Draconetta there is a triangular 



Vol. XXXVIII] GOSLINE: GOBIESOCIFORMES 375 



I'r^^* 




FiGTJRE 6. Draconetta acanthopoma. Pectoral girdle of right side, except postcleithra. 
Lateral line canal indicated by dashed lines, ac, Actinost; cl, cleithrum; co, coracoid; et, 
lateral extrascapular ; li, ligament to intercalar ; sc, scapula ; si, supracleithrum ; and tm, post- 
temporal. 



intercalar on the lower surface of the cranium which serves for the attachment of 
the hgament from the short lower wing of the posttemporal. In Gobiesox and 
Callionymus there is neither an intercalar nor a lower wing to the posttemporal. 

The exoccipital condyles in Draconetta, Callionymus, and Gobiesox are widely 
separated from one another and indeed are practically or quite lateral to the 
basioccipital condyle. As Starks (1905, p. 293, footnote 1) has noted, this con- 
figuration of the occipital condyles is one frequently associated with a depressed 
body form. 

Pectoral girdle. In gobiesocids the supracleithrum and posttemporal bones 
are both present. The cleithrum and primary pectoral girdle extend up the sides 
of the body. From an articulation on the top of the cleithrum, the supracleithrum 
extends horizontally forward, and from the front of the supracleithrum the post- 
temporal extends horizontally inward to an articulation with the skull. The axes 
of the cleithrum, supracleithrum, and posttemporal thus lie primarily in three 
different planes (Guitel, 1889, pi. 24, fig. 3). In Draconetta the supracleithrum 
and posttemporal are present (fig. 6) but the supracleithrum and cleithrum have 
the same axes. Among callionymids Briggs and Berry (1959) and Ochiai (1963) 
state that a supracleithrum and supratemporal are both present, though the 



376 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

latter author shows only one of these two bones in his figures. Starks (1923, p. 
268) says that the supracleithrum is absent in Callionymidae and I can find none 
in Callionymus jlagris, C. decoratus, or Pogonymus. Judging from the position of 
the supracleithrum in Draconetta, it would seem to have become fused with the 
cleithrum in the callionymids investigated by me. Perhaps its loss as a separate 
element is variable in callionymids. 

In Draconetta there are 4 actinosts. The lower 3 are columnar, but the upper- 
most tapers to a basal point and has its entire upper edge contiguous with the 
scapula (fig. 6). In Callionymus, as in the Nototheniidae, there are only 3 acti- 
nosts, the uppermost of Draconetta having doubtless become fused with the 
scapula. 

In Gobiesox there are not only 4, more or less hourglass shaped, actinosts, 
but the scapula projects around the top of the uppermost in such a way as to 
resemble a fifth, as was noted by Starks (1930, p. 220; see also Guitel, 1889, pi. 
24, fig. 10) . It is very probably a scapular projection of similar sort that provides 
the uppermost fifth "actinost" of the batrachoid fishes. 

A further peculiarity of pectoral girdle structure unique among the Gobiesoc- 
idae is the specialization of the two postcleithra (see Starks, 1905; Guitel, 1889, 
pi. 24, fig. 3 ) . Both of the postcleithra on either side are plate-like. The upper 
is vertically alined and has numerous fimbriae extending from its posterior sur- 
face; it appears to be only ligamentously attached to the main pectoral girdle. 
The lower extends inward from the side of the abdomen and, with its counterpart 
from the opposite side, supports the rear rim of the adhesive disc. I do not know 
of a similar specialization elsewhere in fishes, the postcleithra of Chcimarrichthys, 
for example, being quite normal. In Draconetta there is only a single, long, scimi- 
tar-like postcleithrum with the usual ligamentous attachment to the top of the 
cleithrum. Callionymus has an even longer, thinner postcleithral strut, but it is 
made up of 2 pieces closely bound together where they overlap. 

Pelvic girdle. The pelvic girdle of the Gobiesociformes is short and broad, 
as in many notothenioids. The only peculiarity that I can find is in the flat, 
spatulate pelvic spine of Gobiesox, already mentioned. 

Axial and caudal skeletons. In Draconetta there are 7 abdominal and 16 
caudal vertebrae, including the urostylar centrum. In the Callionymidae, so far 
as is known, there are 7+14 vertebrae. Briggs (1955, p. 9) gives the vertebral 
counts of Gobiesocidae as ranging from 25-54; in Gobiesox the count given by 
Starks (1905, p. 300) is 13 + 19. 

In all the Gobiesociformes the ribs start on the second vertebra. In Draco- 
netta and Callionymus there is only 1 pair of ribs per vertebra. These, in Draco- 
netta, extend out and up away from the abdominal cavity, which suggests that 
they are epipleurals. In Gobiesox the same set of ribs occurs, but from the third 
vertebra on there is another set of ribs extending lateroventrally from the lower 
surface of the main ribs about half way out along their length (Runyon, 1961, p. 



Vol. XXXVIII] GOSLINE: GOBIESOCIFORMES 377 

136 and fig. 27). These supplementary lower ribs are, despite their configuration, 
probably pleural ribs (but see Starks, 1905, p. 301). In the flattened nototheni- 
oid Bembrops there is only a single set of ribs, but these commence on the first, 
not the second vertebra ( for the problem of whether a single set of ribs in acan- 
thopterans is pleural or epipleural, see Starks, 1923, p. 290). 

In the gobiesociform fishes examined there are no predorsal bones, and the 
first interneural extends down behind the second neural arch. In Draconetta the 
neural arch to interneural relationship is normal, but in Callionymus and Pogo- 
nymus the third and following vertebrae have V-shaped neural processes that 
extend out laterodorsally on either side of the interneurals. 

In Draconetta there are 2 separate hyjDurals in the caudal skeleton, the lower 
autogenous, and the upper fused to the urostylar centrum. In Callionymus and 
Gobicsox these 2 hypurals are fused into a single unit basally. In Draconetta and 
Callionymus there are 2 epurals, in Gobiesox none. Unlike Gobiesox, the penulti- 
mate vertebra of Callionymus and Draconetta has expanded, plate-like neural 
and hemal arches which are fused to the centrum. 

So far as the fishes examined are concerned, the characters described above 
may be grouped as follows. It should be noted, however, that the wider the spec- 
trum of variation within the group the less any definition based on one or a few 
species, such as those given below, can be expected to hold. 

Gobiesociformes. — Head and body scaleless. Circumorbital bones represented 
only by the lacrimal. Premaxillary with its articular process absent or merged 
with the ascending process (in Draconetta) . Opercular apparatus with 1 or 2 
backwardly projecting spines (except some Gobiesocidae). Metapterygoid 
absent. Ribs commencing on second vertebra. 

Gobiesocoidei. — An abdominal adhesive disc. No spinous dorsal fin. None of the 
fin rays branched. Outer pelvic ray flattened and spatulate, followed by 4 
segmented rays. Palatine separated by membrane from the ectopterygoid. 
No basibranchials. A single upper pharyngeal tooth plate on each side. 
Frontals separate. Mesethmoid not forming part of the orbital boundaries. 
Parasphenoid expanded below and between the orbits. Postcleithra expanded, 
platelike, the lower supporting the rear border of the adhesive disc. ]\Iore 
than 10 abdominal vertebrae, more than 24 in all. Two sets of ribs from the 
third vertebra. Penultimate vertebra with its neural and hemal arches not 
expanded. No epurals. 

Gobiesocidae. — Lateral line system limited to the head. A nasal bone on each 
side of head. Two nostrils on either side, which lead into a nasal sac contain- 
ing a well-developed olfactory rosette. A single spine, if any, on the opercular 
apparatus, formed by the subopercle. Gill openings not restricted to a small 
hole above or behind the opercle. ]\Iesopterygoid absent. Supratemporal 
commissure lacking. No median supraoccipital or frontal crest. Pleurosphe- 
noid, basisphenoid, and intercalar absent. Posttemporal present. Four acti- 



378 CALIFORNIA ACADEMY OF SCIENCES LProc. 4th Ser. 

nosts. Two postcleithra. Neural arches normal. Hypurals fused into a single 
plate. 

Callionymoidei. — No abdominal adhesive disc. A spinous dorsal fin present, 
except Draculo. At least 1 soft ray in each fin branched or divided to the 
base. Outer pelvic ray spinous, followed by 5 soft rays. Palatine firmly at- 
tached to the ectopterygoid. Basibranchials present. Two upper pharyngeal 
tooth plates on each side. Frontals fused or mostly so. Mesethmoid forming 
part of the orbital boundaries. Parasphenoid forming a narrow strut below 
and between the orbits. Postcleithral strut narrow. Seven abdominal verte- 
brae, fewer than 24 in all. A single set of ribs. Penultimate vertebra with its 
neural and hemal arches expanded and plate-like. Two epurals. 

Draconettidae. — Lateral line system limited to head. No nasal bone. Two nos- 
trils on each side of head; no nasal rosette. Two spines on the opercular 
apparatus, one on the opercle and one on the subopercle. Gill openings not 
restricted to a small hole above or behind the opercle. Mesopterygoid present. 
Supratemporal commissure incomplete. A low supraoccipital crest extending 
forward onto the rear of the frontals. Pleurosphenoid, basisphenoid, and 
intercalar present. Posttemporal present. Four actinosts. One postclei thrum. 
Neural arches normal. Two separate hypurals. 

CaUionymidae. — Lateral line continued on body. A nasal bone on each side. One 
nostril leading into a nasal sac with a well developed olfactory rosette. Spine 
on the opercular apparatus single, formed by the preopercle. Gill openings 
restricted to a small hole above or behind the opercle. Mesopterygoid absent. 
Supratemporal commissure complete. No median crest on supraoccipital or 
frontals. Pleurosphenoid, basisphenoid, and intercalar absent. Posttemporal 
absent. Three actinosts. Two postcleithra. Neural arches of third and suc- 
ceeding vertebrae with V-shaped flanges. Hypurals fused into a single plate. 

Of the developments which characterize the Gobiesoci formes as a whole, some 
are of a type that have repeatedly occurred in higher acanthopterans, e.g., the 
"simplification" of skull and fin ray structure. Perhaps the absence of scales and 
the loss of the circumorbital bones behind the lacrimal should be placed in the 
same category. In my opinion the definitive peculiarities held in common by the 
various members of the Gobiesociformes are those of the upper jaw, gill cover, 
and rib configuration. 

That the various members of the Gobiesociformes have diverged widely is 
obvious. In the first place, there is a most remarkable difference in habitat be- 
tween the callionymoids, which are mostly quiet water bottom fishes, and many 
gobiesocids. At least some of the latter, including the close relatives of one dis- 
sected here, live among the boulders of wave-washed rocky beaches. 

The way in which the gobiesocids have evolved from a proto-gobiesociform 
ancestor is suggested by the notothenioid Chcimarrichthys, which has the same 



Vol. XXXVIII] GOSLINE: GOBIESOCIFORMES 379 




GOBIESO CIFORMES 



PERCI 



FORMES 



NOTHENIOIDAE 

Figure 7. Suggested gobiesociform relationships. 

sort of broad, flat head, small, wide-set eyes, and incipient adhesive organ on the 
abdomen as Gobiesox. However, the gobiesocids, in addition to having the pre- 
maxillary, opercular, and rib structure, etc., of all the gobiesociformes, which 
Cheimarrichthys does not have, have incorporated the postcleithra into the ad- 
hesive disc in a unique way. In short, the gobiesocids are much more highly 
specialized fishes than Cheimarrichthys. 

The callionymoids would seem to have diverged from their proto-gobiesoci- 
form ancestors in two principal respects. One is that the high-set eyes have left 
little room for the interorbital portion of the cranium. The frontals have not only 
fused, but their anterior portion appears to have been pinched off and replaced in 
part by the mesethmoid from the preorbital region. Second, there has been a 
reduction in the number of vertebrae. 

Between the draconettids and callionymids, the quite different opercular 
specializations of the two groups preclude the possibility of the one group having 
evolved directly from the other. In general, however, the draconettids have re- 
mained at a lower stage of specialization than the callionymids as indicated by 
the much lower degree of fusion in the draconettid skeleton. 

In my opinion then, the relationships between the three gobiesociform groups 
may be diagrammed as in fig. 7. 

REFERENCES 

BOULENGER, G. A. 

1904. Fishes (systematic account of the Teleostei). In "Cambridge Natural History," 
vol. 7, pp. 541-727, figs. 325-440. 



380 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

Briggs, J. C, and F. H. Berry. 

1955. A monograph of the dingfishes (Order Xenopterygii). Stanford Ichthyological 
Bulletin, vol. 6, pp. 1-224, 114 figs. 
Briggs, J. C, and F. H. Berry. 

1959. The Draconettidae — A review of the family with the description of a new species. 

Copeia, 1959, pp. 123-133, 7 figs. 
Davis, W. P. 

1966. A review of the dragonets (Pisces: Callionymidae) of the western North Atlantic. 
Bulletin of Marine Science, vol. 16, pp. 834-862, 10 figs. 

GOSLINE, W. A. 

1968. The suborders of perciform fishes. Proceedings of the United States National Mu- 
seum, vol. 124, pp. 1-78. 
Greenwood, P. H., D. E. Rosen, S. H. Weitzman, and G. S. Myers. 

1966. Phyletic studies of teleostean fishes, with a provisional classification of living 
forms. Bulletin of the American Museum of Natural History, vol. 131, pp. 
339-456, pis. 21-23, text figs. 1-9. 
Guitel, F. 

1889. Recherches sur les lepadogasters. Archives de Zoologie Experimcntale et Generale, 
ser. 2, vol. 6, pp. 423-627, pis. 24-37. 
Jordan, D. S. 

1923. A classification of fishes, including families and genera as far as known. Stanford 
University Publications, University Series, Biological Sciences, vol. 3, pp. 79-243. 
Kayser, H. 

1962. Vergleichende Untersuchung i.iber Vorstrecksmechanismen der Oberkiefer bei Fis- 

chen. Der Bau und die Funktion des Kiefer- und Kiemenapparates von Kno- 
chenfischen der Gattungen Ammodytes und Callionymus. Zoologische Beitrage, 
new ser., vol. 7, pp. 321-445. 
Linnaeus, C. 

1758. Systema Naturae, Ed. X. Vol. 1. Holmiae: 824 pp. 
McAllister, D, E. 

1968. Evolution of branchiostegals and classification of teleostome fishes. National Mu- 
seum of Canada, Bulletin 221, xiv-f239 pp., 21 pis. 
MONOD, T. 

1960. A propos du pseudobrachium des Antennarius (Pisces, Lophiiformes). Bulletin de 

rinstitut Franqais d'Afrique Noire, vol. 22, ser. A., pp. 620-698, 83 figs. 
OCHIAI, A. 

1963. Two dragonet fishes obtained from the tidal zone of the Amami Islands. Bulletin 

of the Misaki Biological Institute, no. 4, pp. 63-74, 8 figs. 
Regan, C. T. 

1912. The classification of the teleostean fishes of the order Pediculati. Annals and 

Magazine of Natural History, ser. 8, vol. 9, pp. 278-289, 6 figs. 

1913. The classification of the percoid fishes. Annals and Magazine of Natural History, 

ser. 8, vol. 12, pp. 111-145. 
1929. Fishes. In "Encyclopaedia Brittanica," 14th ed,, pp. 305-329. 
Run yon, S. 

1961. Early development of the clingfish, Gohiesox stnimosits Cope. Chesapeake Science, 

vol. 2, pp. 113-141, S3 figs. 

S0RENSEN, W. E. 

1884. Om lydorganer hos fiske. Inaugural dissertation, KJ0benhavn: viii-(-245 pp., 4 pis. 



Vol. XXXVIII] GOSLINE: GOBIESOCIFORMES 381 

Starks, E. C. 

1905. The osteology of Caularchus macandricus (Girard). Biological Bulletin, vol. 9, 
pp. 292-303, 2 figs. 

1923. The osteology and relationships of the uranoscopoid fishes. Stanford University 
Publications, University Series, Biological Sciences, vol. 3, no. 3, pp. 259-290, 
5 pis. 

1926. Bones of the ethmoid region of the fish skull. Stanford University Publications, 
University Series, Biological Sciences, vol. 4, no. 3, pp. 139-338, 58 figs. 

1930. The primary shoulder girdle of fishes. Stanford University Publications, Univer- 
sity Series, Biological Sciences, vol. 6, no. 2, pp. 149-239, 38 figs. 

VAN DOBBEN, W. H. 

1937. tJber den Kiefermechanismus der Knochenfische. Archives Neerlandaises de Zoo- 
logie, vol. 2, pp. 1-72, 50 figs. 



PROCEEDINGS 

OF THE 

CALIFORNIA ACADEMY OF SCIENCES 

FOURTH SERIES 



Festschrift for George Sprague Myers 



Vol. XXXVIII, No. 20, pp. 383-390 December 31, 1970 



SCALE-EATING AMERICAN CHARACOID 

FISHES, WITH SPECIAL REFERENCE TO 

PROBOLODUS HETEROSTOMUS 

By 

Tyson R. Roberts 
Museum of Comparative Zoology, Harvard University, Cambridge, Massachusetts 02138 

INTRODUCTION 

Specialized scale-eaters have been discovered in three groups of American 
characoids. Kner (1860, p. 34) reported the stomach of a large specimen of Cato- 
prion full of scales and referred to them as "Raubfische." Ladiges, observing 
this peculiar serrasalmid in an aquarium, saw one remove a row of scales from a 
specimen of Metynnis with one swipe of its teeth (reported by Gery, 1964, p. 
460). Breder (1927, p. 127) reported substantial amounts of large scales in 
stomach contents of representatives of Roeboides occidentalis from eastern Pan- 
ama and identified some of the scales as coming from Ctenolucius, a pike-like 
characoid considerably larger than the specimens of Roeboides. Gery (1964, pp. 
459-460) reported scale-eating in Exodon, Roeboides, and Roeboexodon, of the 
characid subfamily Characinae. In this paper the activity is verified for Cato- 
prion, Exodon, Roeboexodon, and two additional species of Roeboides and is re- 
ported for the first time in Probolodus heterostomus Eigenmann,^ a member of 
the characid subfamily Tetragonopterinae. 

Although Gery supposed that scale-eating occurred in Catoprion, Exodon, 
Roeboexodon, and Roeboides only occasionally, in these genera and in Probolodus 
scales are definitely a major item in stomach contents, and eating scales is prob- 



1 Myers (1942, p. 91), in recording specimens from the western end of the coastal plain of Rio, com- 
mented on their almost unbelievably strange dentition. The species also occurs in the rios Doce, Paraiba 
(formerly spelt Parahyba), and Ribeira. 

[383] 



384 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

ably a significant factor in the mode of life of these fishes. Serrasalmus elongatus 
includes some scales in its diet but is primarily a fin-eater. Matthes (1961) re- 
ported that the African characoids Phago, Belonophago, and Eugnathichthys 
(family Ichthyboridae) are fin-eaters. There are reports in the literature on 
aquarium fishes that the eating of fins and scales occurs in Phago. Matthes 
(1964, pp. 65-66) reported scales in stomach contents of specimens of Belono- 
phago hutscbouti, Phago boulengeri, and (1961, p. 79) Eugnathichthys, but 
these fishes are primarily fin-eaters. The only other fresh-water scale-eaters of 
which I am aware are certain highly specialized African cichlids (see Fryer^ 
Greenwood, and Trewavas, 1955). I have not searched thoroughly for accounts 
of marine scale-eating fishes. There probably are some; Springer and Woodburn 
(1960, p. 22) stated that annelids and fish scales (with no other fish remains) 
constituted the major portion of stomach contents of sea catfish {Galeichthys 
jelis) taken in Tampa Bay. 

This paper was prepared at the Departamento de Zoologia of the Secretaria 
da Agricultura in Sao Paulo during a visit in April and May, 1969. All obser- 
vations are based on specimens in the Departamento's collections. Measurements 
of fishes given in mm. refer to standard length. 

OBSERVATIONS 
Probolodus. 

Material examined. DZSP 7903, 40 specimens, 41-97 mm., from Rio Pa- 
raiba below represa de Santa Branca (state of Sao Paulo), col. 10-13 February 
1962 by H. A. Britski; and DZSP 7904, 92 specimens, 38-71 mm., represa de 
Santa Branca, Rio Paraiba, collected 10-16 September 1963 by H. A. Britski 
and J. Rossi. 

Stomach contents. Stomach contents were examined in 30 specimens rang- 
ing from 38 to 97 mm. Twenty of these were from the February, 1962 collection, 
and 10 from September, 1963. The stomachs contained food in all specimens. 
Scales were by far the major item encountered and occurred in all but 1 specimen. 
They were the only item present in about 50 percent. The number of scales in a 
stomach varied from 3 to 40, with a mean of about a dozen. Most of the scales 
were 3-5 mm. in diameter, substantially larger than Probolodus^ own scales. A 
white substance of loose consistency was present in large quantity in 5 specimens 
of the February, 1962, sample. Otherwise food items in the 2 samples were very 
similar. The following items were also encountered: small seeds (1 or 2), in 3 
specimens; soil? (small quantities), 3 specimens; minute crustaceans (about 50), 
1 specimen; insect larva (1), 1 specimen; hymenopteran (1), 1 specimen. The 
smallest specimens examined — 38, 41, 47, and 49 mm. — have stomach contents 
similar to the others. 

Dentition. The teeth of Probolodus have been described and figured by 
Eigenmann (1915, pp. 20-21, fig. 5). Probolodus has very few teeth and, as in 



Vol. XXXVIII] ROBERTS: SCALE-EATIXG FISHES 385 

many other characoids with highly specialized dentition, the number is constant 
or very nearly so. Basically there are 3 widely separated teeth on each pre- 
maxillary and 5 on each dentary. Often a tooth is missing, but this is due to loss 
or shedding to make way for a replacement tooth. There are usually either 3 or 
4 teeth on each maxillary, but as few as 2 or as many as 5 were present on some 
sp)ecimens. Here, too, replacement affects the number present. All of the teeth 
are strictly tricuspid. The 3 cusps form a triangle with the enlarged median 
cusp at the anterior angle. The lateral cusps are equal in size and very small. 
The tooth base is moundlike and stout. 

The premaxillary teeth point out of the mouth. The first 3 dentary teeth also 
point out. Only the fourth and fifth dentary teeth lie inside the mouth. The 
enlarged fourth dentary tooth is situated internally to the third and sHghtly 
posterior to it, and the reduced fifth is directly behind the fourth. (Note. — 
Eigenmann refers to one or more small teeth behind the fourth. In specimens I 
have examined there is only one. Perhaps the presence of additional teeth in an 
occasional specimen is a primitive or vestigial character.) The cusps of the ante- 
riormost premaxilliary and dentary teeth point almost straight ahead of the fish. 
The third dentary tooth, and to a lesser extent the third premaxillary tooth, pro- 
ject laterally from the mouth. The teeth are not juxtaposed but are separated 
from each other by a gap about equal to the diameter of a tooth base. WTien the 
mouth is closed the teeth of the upper and lower jaws interdigitate rather than 
truly oppose each other. Thus the first dentary tooth occupies the gap between 
the first and second premaxillary teeth, the second dentary tooth that between 
second and third premaxillary teeth, and the third dentary tooth that between 
the third premaxilliary and first maxillary teeth. The fourth and fifth dentary 
teeth do not oppose or interdigitate with other teeth and neither do the lowermost 
teeth on the maxillary. One can easily imagine how scales are firmly grasped by 
such teeth and then dislodged by the kind of tugging movements many characids 
make when feeding. The number, form, and arrangement of the teeth are the 
same in specimens from 38 to 97 mm. 

Tooth replacement. Twenty specimens from the September, 1963, collec- 
tion were examined for signs of tooth replacement. In only 2 specimens were all 
of the premaxilliary and dentary teeth in functional position and firmly attached 
to the jawbones. In each of the remaining 18 from 1 to 4 teeth were in the proc- 
ess of replacement or had just come into functional position (teeth in the process 
of replacement can be detected immediately below the gum or making their way 
through it; teeth that have just come into functional position are recognizable as 
such because the cusps are unworn and very sharp, the bases are usually sur- 
rounded by soft, swollen tissue, and the attachment to the jaw is very loose) . The 
data indicate that replacement occurs more frequently in lower jaw teeth than in 
upper, and that certain teeth are replaced with relatively high frequency. In all, 
40 instances of tooth replacement in process and teeth newly in functional posi- 



386 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

tion were observed, 29 in the dentary and 11 in the premaxillary. No fewer than 
9 instances involve the fourth dentary tooth. At the other extreme, the first pre- 
maxillary tooth is involved in only 1 instance. Judging from their appearance 
the teeth in the lower jaw receive more wear than those in the upper jaw. 

Sex. Almost all specimens in the February, 1962, collection have readily 
identifiable gonads. The others were sexed by the presence (males) or absence of 
tiny serrations on the anterior anal fin rays. The reliability of this method was 
checked in specimens in which the sex of the gonads was obvious. Of the 40 
specimens in the sample, 29 (72 percent) are females and 11 (28 percent) males. 
Females range from 40.5 to 97 mm. and average 68 mm., 12 mm. more than the 
males. Males range from 47 to 65 mm. and average 56 mm. The largest female 
is 32 mm. longer than the largest male. Combined biomass of females is slightly 
more than three times that of males. The 40.5 mm. specimen contained about 
200 eggs, most .6-. 7 mm. in diameter but a few somewhat smaller; a 73.5 mm. 
specimen contained about 2500, all about .7-. 8 mm. in diameter. 

Serrasalmus elongatus. 

Stomach contents were examined in 7 specimens of S. elongatus Kner, 89-152 
mm., from 3 Amazonian localities. Pieces of the rayed portion of fins and scales 
were present in every specimen ; they were the only items encountered in 5 of the 
specimens. In all but 2 fin rays were by far the major item. One specimen had 
about 50 scales and only a few small bits of fin rays. The 152 mm. specimen (col- 
lected in Lago Jacupa, near Oriximina, state of Para, in February, 1967) contained 
6 cichlid larvae of about 8 mm.; 13 fish? eggs of about 2 mm. in diameter; 2 large 
pieces of very hard, thick fin rays, perhaps from the caudal fin of a sorubim cat- 
fish; and 8 scales about 5-6 mm. in diameter. One specimen, with its stomach 
moderately full of fin rays and a few scales, had a small matted ball of fibrous 
plant material including 3 small seeds. All items encountered have been indicated; 
noteworthy is the absence of pieces of meat. Stomach contents of several Pygo- 
centrus-type piranhas have been examined and when scales were encountered 
there were also bits of meat. Many piranhas feed to some extent on fins. S. 
elongatus is apparently a fin-eater which feeds to a certain extent on scales. 

Catoprion. 

Kner (1860, p. 34) found the stomach of a large Catoprion specimen full of 
scales. Gosline ( 1951, p. 54) examined the stomachs of 4 specimens and reported 
that "two were full of fish scales and two were empty except for a few fish scales; 
a small amount of unidentifiable debris was also found." Gery (1964, p. 460) 
found scales in stomachs of specimens from Bolivia. In examining 4 specimens, 
103-109 mm., from 3 Amazonian localities I find that their stomachs are more 
or less full of scales about 6-15 mm. in diameter. The only other items are 
a few bits of leaf from a higher plant (in 2 specimens) and a small ball of 



Vol. XXXVIII] ROBERTS: SCALE-EATING FISHES 387 

fibrous plant material, probably roots (in 1 specimen). Scales are thus the only 
item that has been encountered in substantial amounts in stomachs of Catoprion. 
The teeth in this genus are illustrated by Miiller and Troschel (1845, pi. 2, 
fig. 5). 

Exodon. 

Gery (1964, p. 459) reported scales in stomach contents of Exodon from the 
Rio Araguaia. I examined 10 specimens, 36-59.5 mm., from the Rio Araguaia 
at Aruana and found from 4 to 15 scales, mostly 3-5 mm. in diameter, in every 
one. The only other item was small amounts of unidentifiable material in 2 speci- 
mens. Kner (1860, p. 47) found beetles in 2 specimens from the Rio Branco. 
The teeth of Exodon are figured by Miiller and Troschel (1845, pi. 4, fig. la). 

Roeboexodon. 

This genus has hitherto been known only from a few specimens taken in 
French Guiana (Gery, 1959). In September, 1966 Heraldo A. Britski and P. E. 
Vanzolini collected 2 specimens (DZSP 4815, 41.5 and 45.5 mm.) from the Rio 
Araguaia near Aruana in the Brazilian state of Goias. The dentition of these 
specimens is identical with that in an alizarin preparation of a 29 mm. specimen 
from French Guiana (kindly sent to the Departamento de Zoologia by Gery) 
and they apparently represent the same species. The stomach contents of both 
specimens consist exclusively of scales from about 2.5 to 4 mm. in diameter. The 
41.5 mm. specimen contained about 10 scales and the 45.5 mm. specimen about 
20. The teeth of Roeboexodon are described and partially figured by Gery (1959, 
pp. 347-349, fig. 2). 

Roeboides. 

Naercio Menezes and I examined stomach contents in 9 specimens of Roebo- 
ides guatemalensis, 6 oi R. myersi, and 25 oi R. prognathus. In all 6 specimens 
of R. myersi (117-160 mm.) and in the 11 largest of R. prognathus (70-90 mm.) 
the stomachs are more or less filled with scales, to the exclusion of all else, those 
of R. myersi with from 15-35 scales mainly 6-9 mm. in diameter and those of 
R. prognathus with 40-150 scales 3-6 mm. in diameter. In 14 smaller examples 
of R. prognathus (41-68 mm.) scales predominate, but insects — Diptera, Hemip- 
tera (Notonectidae?), and a few Coleoptera — occur with high frequency. A 64 
mm. specimen contained a fish larva. Our specimens of R. guatemalensis (72.5- 
101 mm., from Gatun Lake, Panama Canal Zone, collected in November, 1965) 
have viscera heavily infested with nematodes and may not have been feeding 
normally. The stomachs are empty in 4 of them and the other five contain 
but little food, as follows: a few scales (in 4); shrimp (in 2); insect (in 1); 
and an unidentified, flocculent, white material (in 1). 

In very small specimens of Roeboides (20-30 mm.) the teeth can be recognized 



388 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

as belonging to Rocboidcs because of their slightly hypertrophied bases, but they 
are all normal in position. Examination of stomach contents in a few specimens 
(unidentified to species) indicates that at these sizes they feed primarily on 
insects. Only at about 30 60 mm., depending on the species, is the transition 
made to the adult condition in which teeth with greatly hypertrophied bases pro- 
ject from the front of the jaws. 

DISCUSSION 

Stomach contents of fishes belonging to Catoprion, Probolodus, Exodon, Roe- 
boides, and Roeboexodon indicate that their diet consists mainly of scales. The 
teeth are so highly specialized in some of these fishes as to suggest that they 
could not survive in nature on the food that their non-scale-eating ancestors fed 
upon. The remarkable "twin spot" color pattern of Exodon and relatively small 
scales of Probolodus may have evolved after the scale-eating behavior to reduce 
intraspecific scale-eating. Breder (1927, p. 127) speculated that the small, thin, 
and very adherent nature of the scales of Rocboides occidenfalis reduces auto- 
predation. Whereas Catoprion, Exodon, Roeboexodon, and Probolodus are mono- 
typic, Roeboides has speciated extensively. 

The relationships of the five genera, although not yet well understood, show 
that they represent at least three independent lines of evolution: 1. Catoprion 
is definitely a serrasalmid, and probably descended from Serrasalmus. Serrasal- 
mus elongatus includes scales in its diet but is primarily a fin-eater and does not 
appear to be closely related to Catoprion. 2. Eigenmann (1911; 1915) stated 
that Probolodus is very similar in general appearance to Astyanax jasciatus but 
placed it in his polyphyletic subfamily Aphyocharacinae (= Cheirodontinae). In 
my opinion Probolodus belongs in the Tetragonopterinae ; it probably descended 
from Astyanax. It is certainly not related to either Cheirodon or Aphyocharax. 
3. Roeboides is closely related to Charax and Eucynopotamus. Gery (1959, pp. 
404-405) suggested that Exodon was derived from Holobrycon and Roeboexodon 
from Exodon, and placed Roeboides and Charax in a different line. Naercio 
Menezes and I intend to study the osteology of these Characinae in an effort to 
clarify their relationships. We suspect that Eueynopotamus, apparently inter- 
mediate between Charax and Roeboides, is actually based on the young of Roe- 
boides, and note that Roeboexodon bears a strong superficial resemblance to 
Roeboides prognathus Boulenger. 

Perhaps the strange gymnotoid eel Oedemognathus exodon Myers is a scale- 
eater. According to Myers (1936, p. 115), in this apteronotid "the dentigerous 
portion of the premaxillaries is greatly expanded and bulbous, most of it not 
opposable to the lower jaw, and the upper part of it rising above the profile of 
the snout. The whole of this bulbous area is studded with many strong, slightly 
curved, conical teeth, placed irregularly and not very closely together. Most of 
the upper teeth therefore project forward, outward or upward, and are entirely 



Vol. XXXVIII] ROBERTS: SCALE-EATING FISHES 389 

outside the mouth. The lower teeth are similar to the upper ones in shape, and 
are numerous and arranged irregularly, but none is outside the mouth and all 
point in normal direction." Oedemognathus is known only from the holotype, 
202 mm. in total length, USNM 102040, and a 92 mm. specimen reported on and 
figured by Eigenmann and Allen (1942, pp. 325-326, pi. 15, figs. 2-?>), CAS 
(lUM) 15421, both from the Peruvian Amazon. 

ACKNOWLEDGMENTS 

I wish to thank Daniel IVI. Cohen of the U. S. Fish and Wildlife Service and 
Stanley H. Weitzman of the U. S. National Museum for critical comments on the 
manuscript. 

ADDENDUM 

Mr. WiUiam A. Bussing of the Departamento de Biologia, Universidad de 
Costa Rica, informs me that Sr. Carlos Leon, Administrador of the Parque Boli- 
var in San Jose, Costa Rica, observing 2 fish in an aquarium, saw one (Roeboides 
guatemalensis) butt the other {Astyanax sp.) with its snout and then catch the 
dislodged scales as they sank. Bussing has examined the viscera of about 100 
representatives of R. guatemalensis during fieldwork on the Atlantic and Pacific 
slopes of Costa Rica and found almost every specimen had scales and virtually 
nothing else in the stomach. A few contained small insect larvae and one a small 
fish. 

At the John G. Shedd Aquarium in Chicago, Mr, Emanuel Ledecky-Janecek, 
Curator of Exhibits, kindly responded to my request and placed a specimen of 
Leporinus (perhaps L. jriderici) about 9 or 10 inches long in with a small tank- 
ful of fish belonging to Exodon paradoxus. Within a few moments we saw the 
latter agitatedly gang up to one side of the Leporinus victim and take turns 
making extremely rapid circular stabbing motions against its side, always strik- 
ing towards the free margin of the scales, and removing a single scale at about 
every other strike. The scales were swallowed directly. On only one occasion did 
a scale fall to the bottom of the tank and a moment later it too was devoured. In 
about 5 or 10 minutes 20 or 30 scales had been eaten. 

At the Steinhart Aquarium of the California Academy of Sciences I watched 
several fish belonging to Leporinus jasciatus determinedly nipping at fungus- 
infected sores on a specimen of Astronotus ocellatus. I am unsure, but think 
that the cichlid lost a few scales, although the Leporinus specimens seemed to 
confine their nipping to the sores. 

LITERATURE CITED 

Breder, Charles M. 

1927. The fishes of the Rio Chucunaque drainage, eastern Panama. Bulletin of the 
American Museum of Natural History, vol. 57, art. 3, pp. 91-176. 



390 CALIFORNIA ACADEMY OF SCIENCES LProc. 4th Ser. 

ElGENMANN, CaRL H. 

1911. New characins in the collection of the Carnegie Museum. Annals of the Carnegie 

Museum, Pittsburgh, vol. 8, no. ^, pp. 164-181, pis. 4-9. 
1915. The Cheirodontinae, a subfamily of minute characid fishes of South America. 
Memoirs of the Carnegie Museum, Pittsburgh, vol. 7, no. 1, pp. 1-99, pis. 1-17. 
ElGENMANN, Carl H., and W. R. Allen 

1942. Fishes of western South America. University of Kentucky, pp. i-xv, 1-494, map. 
Fryer, Geoffroy, P. H. Greenwood and E. Trewavas 

1955. Scale-eating habits of African cichlid fishes. Nature, vol. 175, no. 4468, pp. 1089- 
90. 
Gery, Jacques 

1959. Contribution a I'etude des poissons characoi'des (Ostariophysi) (II.) Roeboexodon 

gen. n. de Guyane, redescription de R. guyanensis (Puyo, 1948) et relations 
probables avec les formes voisines. Bulletin du Museum National d'Histoire 
Naturelle, second series, vol. 31, no. 4, pp. 345-352; and no. 5, pp. 403-409. 
1964. Poissons characoi'des nouveaux ou non signales de I'ilha do Bananal. Vie et Milieu, 
suppl. 17 (Volume Jubilaire dedie a Georges Petit), pp. 447-471. 
GosLiNE, William A. 

1951. Notes on the characid fishes of the subfamily Serrasalminae. Proceedings of the 
California Academy of Sciences, fourth series, vol. 27, no. 2, pp. 17-61, pis. 1-3. 
Kner, Rudolf 

1860. Zur Familie der Characinen, II. Denkschriften der mathematisch-naturwissen- 
schaftlichen Classe der kaiserlichen Akademie der Wissenschaften, Wien, vol. 
18, pp. 9-62, pis. 1-8. 
Matthes, Hubert 

1961. Feeding habits of some central African fishes. Nature, vol. 192, pp. 78-80. 
1964. Les poissons du Lac Tumba et de la region d'Ikela. Annales Musee Royale de 
I'Afrique Centrale, series in octavo, sciences zoologiques, no. 126, pp. 1-204, 2 
maps, 1 chart, pis. 1-6. 
Muller, Johannes, and F. H. Troschel 

1845. Horae Ichthyologicae. I & II, Die Familie der Characinen. Berlin, pp. 1-40, pis. 
1-11. 
Myers, George S. 

1936. A new genus of gymnotid eels from the Peruvian Amazon. Proceedings of the 

Biological Society of Washington, vol. 49, pp. 115-116. 
1942. Studies on South American fresh-water fishes. I. Stanford Ichthyological Bulletin, 
vol. 2, no. 4, pp. 89-114. 
Springer, Victor G., and K. D. Woodburn 

1960. An ecological study of the fishes of the Tampa Bay area. Florida State Board of 

Conservation, Marine Laboratory (St. Petersburg), Professional Papers Series, 
no. 1, pp. 1-104. 



PROCEEDINGS 

OF THE 

CALIFORNIA ACADEMY OF SCIENCES 

FOURTH SERIES 



Festschrift for George Sprague Myers 



Vol. XXXVIII, No. 20, pp. 383-390 December 31, 1970 



SCALE EATING AMERICAN CHARACOID 

FISHES, WITH SPECIAL REFERENCE TO 

PROBOLODUS HETEROSTOMUS 

By 

Tyson R. Roberts 
Museum of Comparative Zoology, Harvard University, Cambridge, Massachusetts 02138 

INTRODUCTION 

Specialized scale-eaters have been discovered in three groups of American 
characoids. Kner (1860, p. 34) reported the stomach of a large specimen of Cato- 
prion full of scales and referred to them as "Raubfische." Ladiges, observing 
this peculiar serrasalmid in an aquarium, saw one remove a row of scales from a 
specimen of Metynnis with one swipe of its teeth (reported by Gery, 1964, p. 
460). Breder (1927, p. 12 7) reported substantial amounts of large scales in 
stomach contents of representatives of Roeboides occidentalis from eastern Pan- 
ama and identified some of the scales as coming from Ctenolucius, a pike-like 
characoid considerably larger than the specimens of Roeboides. Gery (1964, pp. 
459-460) reported scale-eating in Exodon, Roeboides, and Roeboexodon, of the 
characid subfamily Characinae. In this paper the activity is verified for Cato- 
prion, Exodon, Roeboexodon, and two additional species of Roeboides and is re- 
ported for the first time in Probolodus heterostomus Eigenmann,^ a member of 
the characid subfamily Tetragonopterinae. 

Although Gery supposed that scale-eating occurred in Catoprion, Exodon, 
Roeboexodon, and Roeboides only occasionally, in these genera and in Probolodus 
scales are definitely a major item in stomach contents, and eating scales is prob- 



^ Myers (1942, p. 91), in recording specimens from the western end of the coastal plain of Rio, com- 
mented on their almost unbelievably strange dentition. The species also occurs in the rios Doce, Paraiba 
(formerly spelt Parahyba), and Ribeira. 

[383] 



392 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

anal spine has been overlooked in Naso, a genus thought to be alone among the 
acanthurids in having only two anal spines rather than the normal three. 

Naso, like Prionurus (including Xesurus), has fixed plates on the caudal 
peduncle rather than the folding spines typical of all other acanthurids. Naso 
and Paracanthurus are the only genera with 3 soft rays in the pelvic fin rather 
than 5. On the basis of Naso and Prionurus having fixed peduncular plates, Smith 
(1955: 169) recognized these 2 groups (with Naso divided into 5 genera on the 
basis of the number of peduncular plates and the snout horn development) as the 
family Nasidae distinct from the Acanthuridae. Randall (1955: 365-366), in 
a brief addendum to his revision of the surgeon fish genera, gave excellent rea- 
sons for not accepting this splitting of the family and of the genus Naso. Smith 
(1966: 635) subsequently recognized Naso (this time divided into 3 genera) as 
the subfamily Nasinae on the basis of its having 2 anal spines and 3 pelvic rays, 
in contrast to the 3 anal spines and 5 pelvic rays of the other acanthurids which 
he divided into 2 subfamilies on the basis of whether the caudal peduncle has 
folding spines (Acanthurinae) or fixed plates (Prionurinae). Such a system 
neglects Paracanthurus, which, in addition to 3 anal spines, has folding pedun- 
cular spines and only 3 pelvic rays. In Smith's system, Paracanthurus would 
have to be recognized as an additional subfamily. By most contemporary stan- 
dards this system would seem to be far too finely split at the subfamilial and 
generic levels on the basis of somewhat superficial characters, even if all these 
characters were valid. 

The fact that Naso has 3 anal spines like the other acanthurids is additional 
evidence that this genus should not be considered as a subfamily distinct from 
the other acanthurids. The anal spine that is supposedly absent in Naso is the 
first spine. This spine is substantially similar to the first spine in other acanthu- 
rids, except that its distal ix)rtion which would protrude through the skin is lost, 
leaving only the basal portion which acts as a complex locking device in basically 
the same manner as in other acanthurids. 

An initial survey of acanthurid osteology, which cannot be dealt with here, 
based on representatives of all of the genera recognized by Randall (1955), 
shows no features that warrant the recognition of subfamilies within the group. 

LOCKING MECHANISM OF THE DORSAL AND ANAL SPINES 

The locking mechanism in Acanthurus triostcgus is osteologically representa- 
tive of all of the acanthurids except Naso, and the descriptions and illustrations of 
the bony parts given here based on A . triostegus should apply well except in fine 
detail to the various species of Acanthurus, Ctenochaetus, Paracanthurus, Prio- 
nurus, and Zcbrasoma. The musculature of the locking mechanism is described 
for A. triostegus, and although this has not been compared with the situation in 
the other genera that have 3 obvious anal spines, I suspect that, based on the 
shapes of the bony parts, it is similar in all of them. The bony parts of the 



Vol. XXXVIII] TYLER: SPIXE LOCKING L\ SURGEOX FISHES 393 

locking mechanism as described for A'aso literatus are typical of that genus, 
and the few comments given here on the musculature probably will apply to all 
species of Naso. 

The dorsal and anal spines and their pterygial supports are more generalized 
in Acanthurus than in Naso, and are described first below. 

Acanthurus triostegus (Linnaeus). 

Specimens examined. ANSP 109491, 7 specimens, 40.6-68.9 mm. stan- 
dard length, Caroline Islands, cleared and stained. ANSP 109490, 1, 69.0 mm., 
no locality, cleared and stained. ANSP 108288, 4 (out of 12), 83.6-97.0 mm., 
Seychelles Islands, alcohol. 

Dorsal spines and pterygiophores. The first two are borne on the 
first basal pterygiophore, and the subsequent seven spines on their own individ- 
ual basal pterygiophores as well as distal pterygiophores. The first two spines ar- 
ticulate medially with the large flattened medial flange at the distal end of the 
basal pterygiophore, which projects above the level of the distal ends of the subse- 
quent pterygiophores, as well as ventrolaterally by their divergent bases with lat- 
eral flanges on the sides of the pterygiophore. The third and subsequent spines 
articulate at their less divergent bases with the regions of suturing between the 
distal pterygiophores anteroventral to their bases and the basal pterygiophores 
ventrally and posteroventrally. The first basal pterygiophore is held basally be- 
tween the exoccipitals and the dorsal half of the neural spine of the first vertebra. 
Each of the subsequent basal pterygiophores of the spiny dorsal fin is held between 
the neural spines of adjacent vertebrae, except that there is no pterygiophore 
between the third and fourth neural spines (fig. 1). This is true of all acanthurids, 
regardless of the number of dorsal spines, and is also true of zanclids. In the 
closely related siganids there is one basal pterygiophore of the spiny dorsal fin be- 
tween adjacent neural spines, except that there is no pterygiophore between the 
fifth and sixth neural spines. In siganids the first dorsal spine is normal, not 
modified into an acanthurid-like locking mechanism. However, the distal ends 
of the pterygiophores are laterally expanded into plates in much the same manner 
as explained below for acanthurids, especially Naso. 

The basal region of the first spine is deeply concave, while that of the second 
spine has a complete foramen anteroposteriorly through which the medial flange 
of the first basal pterygiophore passes, the flange in this region having a hole to 
accommodate the extreme ventral end of the second spine, which is solid and 
without a medial suture (figs. 2-?>). The third spine also has a complete antero- 
posterior foramen in its base through which passes the medial bridge formed by 
the posterodorsal process of the first distal pterygiophore and the anterodorsal 
process of the second basal pterygiophore. The fourth and subsequent dorsal 
spines usually have a complete foramen ventrally. although the ventromedial 
region may have a sutural mark medially. The bridges formed by the processes 



394 



CALIFORNIA ACADEMY OF SCIENCES 



[Proc. 4th Ser. 



praocctpi*al 



obdommcl vertebrae 




Figure 1. Lateral view of vertebral column and fin supports in Aranthurus triostegns, 
ANSP 109491, 46.2 mm. standard length. Bases of fin rays indicated in black; distal pteryg- 
iophores of fin rays not shown. 



Vol. XXXMin TYLER: SPIXE LOCKING L\ SURGEOX FISHES 



395 




1st distal ptetygloDhore 



Figure 2. Lateral views of first three dorsal and anal spines and their supports in Acan- 
thuriis triostegus, the spines only partially erected; composite drawings based on specimens 
from ANSP 109491. 



of the distal and basal pterygiophores, around which the foramina of the spines 
articulate, are slightly less well developed posteriorly in the series than anteriorly, 
the anterodorsal process of the basal pterygiophore especially tending to be of 
lesser length so that it does not quite contact the process of the distal pterygio- 
phore in front of it and thus fails to form a complete bridge. 

Distally the pterygiophores are expanded laterally to form a plate composed 
anteriorly of the expanded posterior half of the distal end of the basal pterygio- 
phore and posteriorly by the expanded anterior portion of the distal pterygio- 
phore. The tendinous insertions of the erector muscles of the spines (except for 
those of the first 2 spines) are accommodated by gaps between the composite 
plates, the gaps being between the posterior edges of the expanded portions of 
the distal pterygiophores and the anterior edges of the expanded portions of the 
basal pterygiophores (fig. 4). The depressor muscles are accommodated by 
similar gaps in the anterior third of the laterally expanded plates of each of the 
basal pterygiophores. The anterolateral edges of the basal pterygiophore plates 
tend to be prolonged anteriorly, partially enclosing the lateral surfaces of the 
insertion ends of the depressor muscles. The amount of bridging in these regions 



396 



CALIFORNIA ACADEMY OF SCIENCES 



[Proc. 4th Ser. 




Figure 3. Lateral view of first two dorsal spines in Acanthunis triostegiis, the spines 
almost fully erected; composite drawing based on specimens from ANSP 109491. 



Vol. XXXVIII] TYLER: SPINE LOCKING IN SURGEON FISHES 397 

1st basal pterygiophore 2nd -b^ 3rd 

basal pterygiophores 




distal pterygiophores 



Figure 4. Dorsal view of first three basal and distal pterygiophores of dorsal fin of 
Acanthuriis triostegus, after removal of the spines, ANSP 109491, 46.2 mm. standard length. 



may increase with increasing specimen size, but it probably never reaches the 
complete bridging found in Naso, as discussed later. 

As one of the major characteristics of the acanthurids, Gill stated (1884: 
276; this statement often followed by others): "Interneurals with expanded 
buckler-like subcutaneous plates, which intervene between the spines and limit 
their erection forwards." The plates he refers to obviously are the laterally ex- 
panded distal ends of the basal and distal pterygiophores, but these plates in no 
way interfere with the full erection of the spines, as explained below. Connective 
tissue in dried skeletal preparations or unnaturally inelastic membranes and 
ligaments in alcohol preserved specimens must have misled him. 

All of these spines can be erected at right angles to their pterygiophores. The 
first and second spines are bound together by a compact ligament, while the 
second and subsequent spines are bound together by more diffuse ligamentous 
connections spread out over most of the lengths of the spines. The erection or 
depression of one spine thus is in concert with the others. When the spines are 
fully erected at right angles to their pterygiophores, they tend to stay erected 
because of the frictional resistance of their ventral edges against the dorsal sur- 
faces of their pterygiophores caused by the down pull of the erector muscles, 
but the whole series of spines can be firmly locked in an erected position varying 
from full to any degree of partial erection by a catching action of the small first 
spine against grooved surfaces on the dorsomedial flange of the first basal 
pterygiophore. 

The locking mechanism probably develops very early in the larval stages, 
for the figures of 3 to 5 mm. long specimens of A. monroviae given by Abousso- 



398 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

uan (1965: figs. 2-3) clearly indicate the grooved medial flange of the first 
basal pterygiophore, although the small first spine is not shown. 

The dorsomedial flange of the first basal pterygiophore is generally circular 
in outline as seen laterally, with the exception of the region just behind the mid- 
dle of the flange discussed below. The flange bears grooves running from its 
distal edge approximately towards the center. The deeply concave ventral sur- 
face of the first spine is relatively smooth and rides over the grooved peripheral 
surface of the medial flange of the first basal pterygiophore. The ventral arms, 
or extreme basal flanges, of the first spine rotate around a small flange projecting 
laterally from the lateral surface of the pterygiophore at the level of the horizon- 
tal base of the medial flange. The dorsal surface of this pterygiophore flange 
and the ventral edge of the ventral arm of the first spine which it supports are 
both knurled, the amount of knurling being highly variable in the specimens 
studied. As the first spine rotates in erection forward and downward it slides 
easily over the variously anteriorly to dorsally oriented grooves on the medial 
flange of the basal pterygiophore. When a downward and/or backward pressure 
is exerted on the first spine, it catches firmly on the grooves of the medial flange. 
Since the first and subsequent spines are ligamentously connected, the whole 
spiny dorsal fin is firmly held erect. Because the grooves on the medial flange 
are continuous from the region underlying the concave ventral surface of the 
first spine in its unerected position to its fully erected position at right angles to 
the pterygiophore, the first spine can be locked in any of the innumerable posi- 
tions between the two extremes. The first spine is unlocked from its erected 
position by relaxation of the pressure of the muscles (as discussed below) pulling 
it into intimate contact with the grooves on the medial flange of the basal pteryg- 
iophore, allowing it to slide upward and posteriorly without undue frictional 
resistance. The rotational course of the first spine is blocked anteriorly when it 
reaches its fully erected position at about a right angle to the pterygiophore by 
its anterior edge hitting against two bony obstacles on a prong-like portion of the 
pterygiophore anterior to the medial flange and separated from it by a deep 
vertical canal into which the head of the first spine rotates in erection. The 
extreme anterior end of the first spine hits the bottom of the canal at the same 
time that a slight indentation on the mid-dorsal (as seen when unerected) edge 
of the spine hits against the posterodorsal edge of the prong-like portion of the 
pterygiophore. 

The second spine rotates over the posterior half of the medial flange of the 
basal pterygiophore. The peripheral surface of the flange over which it slides 
is also grooved, although not so deeply or regularly as that portion over which 
the first spine slides. Nevertheless, downward pressure on the second spine 
serves to help lock the spiny dorsal fin in an erected position. The ventral arms 
(which are fully fused in the region of the foramen in the medial flange) of the 
second spine rotate around the oblique dorsal surface of a large flange running 



Vol. XXXVIII] TYLER: SPIXE LOCKING L\ SURGEON FISHES 399 

vertically along most of the lateral surface of the first basal pterygiophore. The 
rotational course of the second spine is blocked anteriorly when it reaches a fully 
erected ix)sition at about a right angle (or a little more) to the pterygiophore by 
its anterior edge just at the dorsal end of the anteroposterior hole through its 
base hitting against an indented and thickened region on the medial flange. The 
medial flange is thicker beneath the position of the base of the unerected first 
dorsal spine than more posteriorly, and the lateral ridge so formed corresponds 
to an indented region on the flange just posterior to the ridge. The anterior edge 
of the second spine in full erection hits against this ridge and indentation, stop- 
ping any further rotational movement forward. 

The third and subsequent spines, by comparison, rotate simply over their 
articulations with the bridges formed by the posterodorsal processes of the distal 
pterygiophores and the anterodorsal processes of the basal pterygiophores, with- 
out any special mechanisms for locking them in an erected position, other than 
their ligamentous connections with the first 2 spines. 

Musculature. When the skin is removed from the upper part of the trunk 
below the dorsal spines, 3 main muscle groups are seen (fig. 5, top) : the general 
epaxial muscle mass; the inclinators which originate on the undersurface of the 
skin and course posterodorsally to insert variously on the expanded plates of the 
pterygiophores and on the ventrolateral surfaces of the bases of the spines; and 
the supracarinales anterior, a band of longitudinal muscle running from the 
supraoccipital to the lateral surfaces of the anterodorsal prong-like portion of 
the first basal pterygiophore and of the base of the first spine. The epaxial 
muscle mass has nothing to do with the erection mechanism and is not further 
discussed. The inclinators of the first 2 spines are poorly developed in compari- 
son to those of the subsequent spines and of the soft rays, and they probably 
have little bearing on the locking mechanism. The band of muscle from the 
supraoccipital divides posteriorly into a deep segment attaching to the prong 
of the basal pterygiophore and a superficial segment attaching to the basal region 
of the first spine. The segment attached to the pterygiophore prong apparently 
is not associated with the locking mechanism, but the segment attached to the 
spine may help to unlock the spine prior to depression of the fin. When the first 
spine has been pulled firmly against the grooved surface of the pterygiophore 
flange it is locked in place, and a forward and/or upward directed pressure is 
necessary to disengage it from the grooves before it can slide without undue 
frictional resistance over the flange. Contraction of the longitudinal band of 
muscle between the supraoccipital and the first spine probably helps to disengage 
the spine by pulling it forward. However, the first anal spine has the same type 
of locking mechanism as the first dorsal spine, yet the anal spine does not have a 
band of muscle running forward from its base which, when contracted, would 
tend to disengage the spine from its locked position, nor is such a muscle present 
on the first dorsal and anal spines of Naso. I assume that the natural elasticity 



400 



CALIFORNIA ACADEMY OF SCIENCES IProc. 4th Ser. 



1st basal pterygiophore 2nd basal pterygiophore 

2nd spine / 3rd spine 




2nd distal pterygiophore 
2nd basal pterygiophore 



1st abdominal vertebra 



Figure S. Lateral views of musculature of first three dorsal spines in Acanthiirus trioste- 
giis; above, superficial musculature after removal of skin; below, erector and depressor 
muscles as seen after removal of superficial muscles: composite drawings based on specimens 
from ANSP 108288; the representational fiber structures are diagrammatic. Legend: A, 
supracarinales anterior; B, inclinators; C, epaxial muscle mass; D, depressors; E, erectors. 



and resiliency of the various connective tissues between the first spine and the 
pterygiophore are strong enough to push or pull the spine the slight distance 
necessary to relieve the pressure on its concave ventral surface from the grooves 
on the medial flange of the pterygiophore. The relative smoothness of the ventral 



Vol. XXXVIII] TYLER: SPINE LOCKING IN SURGEON FISHES 401 

surface of the spine perhaps facilitates the unlocking of the mechanism, for cer- 
tainly if it were deeply grooved like the medial flange a complex mechanism for 
unlocking the adpressed grooved surfaces would have to be present. 

The erector muscle of the first spine is the largest in the series (fig. 5, bot- 
tom). It originates over the middle of the posterior wall of the skull and over 
the lateral surfaces of the first vertebral neural spine and first pterygiophore. It 
inserts on the forward half of the lateral face of the base of the spine. In the 
upper third of its length, the muscle is partially divided into a forward and rear 
segment. Contraction of the muscle pulls the spine forward and downward over 
the medial flange. When the spine is erected it is probably the contraction of 
the rear segment of the upper third of the muscle that pulls the spine into close 
contact with the grooves of the medial flange. The muscle overlies much of the 
length of the large lateral flange running vertically on the pterygiophore, the 
head of this flange being the point around which the second spine pivots. The 
depressor muscle of the first spine originates under the upper regions of the 
erectors of the first and second spines. It is much smaller than any of the other 
depressor muscles in the series. It inserts through a band of connective tissue on 
a posteriorly directed process from the posterior edge of the basal flange of the 
first spine. The contraction of the depressor muscle simply rotates the first spine 
backward over the medial flange to its unerected position. 

The erector and depressor muscles of the second and subsequent spines re- 
quire no special comment. 

Anal spines and pterygiophores. The 3 anal spines and the distal jwr- 
tions of the pterygiophores to which they articulate have much the same shapes 
and relationships as the first 3 dorsal spines do to their pterygiophores, as de- 
scribed briefly above. The ventral surface of the basal half of the first spine is 
deeply concave, fitting over a grooved medial flange of the first basal pterygio- 
phore; the second and third spines have complete anteroposterior foramina 
through their bases, that of the second spine fitting through the bridging 
of a hole in the posterior half of the medial flange of the first pterygiophore and 
that of the third spine through the bridge formed by the posteroventral process 
of the first distal pterygiophore and the anteroventral process of the second basal 
pterygiophore. The third anal spine is stouter than the third dorsal spine, and 
the distal ends of the pterygiophores are not expanded into large plate-like struc- 
tures such as are found on the pterygiophores of the spiny dorsal fin. The lock- 
ing mechanism of the first anal spine is like that of the first dorsal spine, and the 
rotational course of the first 2 spines is blocked when the spines are about at a 
right angle to the pterygiophore in the same manner as described for the spiny 
dorsal fin. 

The long shaft-like proximal portion of the first basal pterygiophore is firmly 
held by connective tissue against the anterior edge of the haemal spine of the 
tenth vertebra, the first of the caudal series. The proximal portion of the second 



402 



CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 



haemal spine- lOlh vertebra 



2nd basal pterygiophore 




pelvis 



2nd spine 



Figure 6. Lateral views of musculature of anal spines in Acanthunis triostegus; below, 
superficial musculature after removal of skin; above, erector and depressor muscles as seen 
after removal of epaxial muscles and inclinators; composite drawings based on specimens 
from ANSP 108288; the representational fiber structures are diagrammatic. Legend: A, 
infracarinales medialis; B, inclinators; C, hypaxial muscle mass; D, depressors; E, erectors. 



basal pterygiophore is held against the posterior edge of this haemal spine, while 
the first and second basal pterygiophores of the soft anal fin are held respectively 
to the anterior and posterior edges of the haemal spine of the second caudal 
vertebra. 

Musculature. Three main muscle groups are apparent when the skin is 



Vol. XXXVIII] TYLER: SPINE LOCKING IN SURGEON FISHES 403 

removed from the trunk above the anal spines (fig. 6, bottom): the general 
hypaxial muscle mass; the inclinators; and a band of longitudinal muscle run- 
ning from the lateral surface of the pelvis just behind the level of the fin rays to 
the lateral surface of the anteroventrally directed medial prong-like portion of 
the first basal pterygiophore. The inclinators apparently are increased in number 
relative to those of the dorsal spines and are not associated with the anal spines 
on a one-to-one basis as they are with the dorsal spines. They originate on the 
undersurface of the skin and insert variously on the distal regions of the pteryg- 
iophores and on the lateral surfaces of the bases of the spines. The inclinators, 
like the hypaxial muscles, have no bearing on the locking mechanism. The band 
of muscle (infracarinales medialis) between the pelvis and first basal pterygio- 
phore becomes tendinous as it passes around the anus. This muscle in some ways 
corresponds to the one in the spiny dorsal fin which runs from the supraoccipital 
to the prong of the first basal pterygiophore and base of the first spine. How- 
ever, in the anal fin this muscle inserts only on the anteroventral prong of the 
first basal pterygiophore and does not have a superficial segment making con- 
tact with the first spine. It apparently has nothing to do with the locking mecha- 
nism. 

The erector and depressor muscles are shown at the top of figure 6. As with 
the first dorsal spine, the erector muscle of the first anal spine is the largest in 
the series, and its depressor is the smallest. The erectors and depressors of the 
subsequent spines are of about equal size. They rotate the spines in the same 
manner as described for the spiny dorsal fin. 

Naso literatus (Bloch and Schneider). 

Specimens examined. ANSP 109497, 2, 111.2-208.5 mm., tropical western 
Pacific, cleared and stained. ANSP 109496, 1, 191.4 mm., no locality, cleared 
and stained. ANSP 108416, 1, 107.3 mm., Seychelles Islands, cleared and stained. 
ANSP 108272, 1, 168.3 mm., Seychelles Islands, alcohol; species undetermined, 
but not N. literatus, used for muscle examination. 

Dorsal spines and pterygiophores. There are 7 dorsal spines, the first 
of which is so distally aborted that it does not show externally, and the dorsal 
spine counts given in the literature are nearly always one less than in actuality 
(fig. 7). The second and subsequent dorsal spines are similar in their articula- 
tions to those described above for Acanthurus triostegus, except that the spines 
of Naso are stouter (and more heterocanth) and the distal pterygiophores to 
which the third and subsequent spines articulate are smaller, the latter difference 
being described below. The first dorsal spine of Naso differs from that of Acan- 
thurus and the other genera of acanthurids only in having lost its distal portion 
that would ordinarily protrude through the skin (fig. 8). The basal region of 
the spine (all that remains of it) is similar to that described above for A. trio- 
stegus, including a deeply concave smooth ventral surface, ventral arms articu- 



404 



CALIFORNIA ACADEMY OF SCIENCES 



[Proc. 4th Ser. 



supraoccipital 




Figure 7. Lateral view of vertebral column and fin supports in Naso litemtus, ANSP 
109497, 111.2 mm. standard length. Bases of fin rays indicated in black; distal pterygiophores 
of fin ravs not shown. 



Vol. XXXVIII] TYLER: SPINE LOCKING IN SURGEON FISHES 



405 



2nd spi ne 




1st basal 
pterygiophore 



2nd spine 



distal pterygiophore 



base of 1st fin ray 



'3rd spine 



Figure 8. Lateral views of first three dorsal and anal spines and their supports in Naso 
literatus, the spines only partially erected; composite drawings based on all of the cleared 
and stained materials listed. 



lating with lateral flanges on the first basal pterygiophore, and a posteriorly 
directed process from the posterior edge of the ventral arm to receive the inser- 
tion of the depressor muscle. A compact ligament connects the first spine at a 
groove along its dorsolateral surface to the anterolateral surface of the basal 



406 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

region of the second spine, and the second and subsequent spines are connected 
by more diffuse ligamentous connections, just as described for .-1. triostegus. 

The major differences between A. triostegus (and the other acanthurids with 
an externally visible first dorsal spine) and N . liter atus (and the other species of 
Naso), other than the smaller size of the first dorsal spine and the somewhat 
stouter size and heterocanth arrangement of the other spines, are in the shapes 
of the pterygiophores supporting them. The medial flange of the first basal 
pterygiophore is much less deeply grooved in Naso than in other acanthurids, 
and the anterodorsal prong of the pterygiophore, which in other acanthurids is 
relatively laterally compressed, in Naso is expanded posterolaterally into a 
bridge which meets and fuses or sutures with the lateral flange on the pterygio- 
phore around which the ventral arms of the first spine rotate. The canal between 
the prong and the rest of the pterygiophore remains, and into this canal the first 
spine rotates when erected, just as described for A. triostegus. The rotational 
course of the first 2 spines is halted at about a right angle to the pterygiophore 
when the anterior edge of the base of second spine hits the indented region on the 
medial flange of the pterygiophore, and when, at the same time, the first spine 
can travel no farther in the canal between the anterodorsal prong and the medial 
flange. The second spine has a complete foramen anteroposteriorly through its 
base, the ventral bridge of the foramen being accommodated in a hole through 
the medial flange, as in A. triostegus. The third and subsequent spines also have 
complete foramina, these articulating around the bridges formed by the postero- 
dorsal processes of the distal pterygiophores and the anterodorsal processes of 
the basal pterygiophores. 

The basal pterygiophores of the spiny dorsal fin in Naso otherwise differ from 
those described in A. triostegus mainly by the greater development of the ex- 
panded plates at their distal ends, and of the lesser involvement in this structure 
of the distal versus basal pterygiophores (fig. 8). In .^. triostegus the distal 
pterygiophores have expanded anterolateral wings which contribute substantially 
to the plates, but in N . literatus the plates are formed only by the basal pteryg- 
iophores, the distal pterygiophores remaining relatively small and medially 
placed in comparison to the width of the plates. Large foramina are present on 
each side of each expanded basal pterygiophore plate; these accommodate the 
depressor muscles of the second and subsequent spines, the erector muscles being 
accommodated in the gaps between the basal pterygiophore plates bordered medi- 
ally by the small distal pterygiophores (fig. 9). 

The locking mechanism in Naso works just as described for A. triostegus, and, 
despite the less deep grooving on the medial flange in Naso, the apparatus seems 
to lock equally firmly. 

Musculature. The muscles of Naso have been examined only cursorily, 
and in a species other than N . literatus. But the major features of the bony struc- 



Vol. XXXVIII] TYLER: SPIXE LOCKING L\ SURGEOX FISHES 



407 



1st basal pterygiophore 



2nd V3rd 
basal pterygiophores 




dorsal spine 



distal pterygiophores 



Figure 9. Dorsal view of first tiiree basal and distal pterygiophores of anal fin of Naso 
Uteratus, with the first spine in place at a position about halfway to full erection and the 
second and third spines removed, ANSP 108416, 107.3 mm. standard length. 



tures in the appropriate regions are similar in all species of Xaso, and the muscu- 
lature probably follows suit. 

The longitudinal band of muscle (supracarinales anterior) between the supra- 
occipital and the anterodorsal prong of the first basal pterygiophore does not 
contact the first dorsal spine, its path being blocked by the posterolaterally ex- 
panded portion of the prong which forms a bridge to the lateral flange support- 
ing the ventral arm of the first spine. The erector and depressor muscles of the 
spines do not seem out of the ordinary, although the depressor muscles are even 
more deeply buried beneath the erectors than the condition described and shown 
above for A . triostegus. 

Anal spines and pterygiophores. The 3 anal spines and associated re- 
gions of the pterygiophores are similar in their relationships to that described 
for the first 3 dorsal spines, except that there is no separate distal pterygiophore 
between the first and second basal pterygiophores, although each of the follow- 
ing soft rays has separate distal pterygiophores. 

The large proximal shaft-like portion of the first basal pterygiophore is 



408 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

firmly held along the anterior edge of the haemal spine of the tenth vertebra, the 
first of the caudal series, and is curved much farther anteroventrally than in A. 
triostegus. The second basal pterygiophore is aborted proximally and squeezed 
in between the posteroventral region of the first basal pterygiophore of the spiny 
anal fin and the anterior edge of the slightly larger first basal pterygiophore of 
the soft anal fin. The second to sixth basal pterygiophores of the soft anal fin 
follow in series between the haemal spines of the first 2 caudal vertebrae. 

Musculature. As in A. triostegus, the muscle (infracarinales medialis) 
between the pelvis and the first basal pterygiophore does not contact the first 
spine and become tendinous as it passes around the anus. The shapes and sizes 
of the erector and depressor muscles of the anal spines are not clear to me in the 
single specimen examined, but the depressor muscles seem even more deeply 
buried below the erectors than in the dorsal fin. 

Miscellaneous notes on the axial skeleton. As indicated in figures 
1 and 7, acanthurids have 3 epurals, the first in close association with the neural 
arch of the penultimate vertebra, and a well-developed paired uroneural. The 
hypurals (including the parhypural) number six in most acanthurids, as shown 
for A. triostegus, but in the several species of Naso investigated, the middle 4 
elements are fused to each other and to the centrum, leaving only 2 separate 
elements. The haemal spines of the penultimate and antipenultimate vertebrae 
are autogenous in acanthurids. 

ACKNOWLEDGMENTS 

Mr. Richard Winterbottom, Queen's University, Kingston, Ontario, Canada, 
generously advised on the musculature of the acanthurid locking mechanism, 
and Dr. Daniel M. Cohen, U.S. Fish and Wildlife Service, Washington, D.C., 
gave useful comments on the manuscript. The halftone illustrations are by the 
extremely competent Mrs. Mary H. Fuges. 

More generally, I am in the debt of Professor Myers for much good counsel 
and encouragement in the study of plectognath fishes and their relatives. This 
short paper in his honor is only an infinitesimal token of my deepest gratitude 
for his innumerable hours devoted to my education. 

LITERATURE CITED 
Aboussouan, a. 

1965. Oeufs et larves de Teleosteens de I'Ouest africain. I. — Acanthurus monroviae 
Steind. Bulletin de ITnstitute Franqais d'Afrique Noire (Dakar), (A), vol. 27, 
no. 3, pp. 1183-1187, 3 figs. 

DONITZ, W. 

1867. Ueber die Gelenke an der Riicken- und Afterflosse der Tcuthics C. Val. Archiv 
fiir Anatomic, Physiologie, und wissenschaftliche Medizin (Leipzig), Jahrbuch, 
1867, no. 1, pp. 210-220, pi. 7a. 
Gill, Theodore N. 

1884. Synopsis of the genera of the superfamily Teuthidoidea (families Tcuthididae and 



Vol. XXXVIII] TYLER: SPIXE LOCKING LN SURGEON FISHES 409 

Siganidae). Proceedings of the United States National Museum (Washington, 
D.C.), vol. 7, no. 18, pp. 275-281. 
GiJNTHER, Albert 

1861. Catalogue of the acanthopterygian fishes in the collection of the British Museum, 
vol. i, London, pp. 1-586. 
Randall, John E. 

1955. An analysis of the genera of surgeon fishes (family Acanthuridae). Pacific Science 
(Honolulu), vol. 9, no. .^ pp. 359-367. 
Smith, J. L. B. 

1955. East African unicorn fishes from Mozambique. South African Journal of Science 

(Johannesburg) vol. 51, no. 6, pp. 169-174, 2 pis. 
1966. Fishes of the sub-family Nasinae with a synopsis of the Prionurinae. Ichthyolog- 
ical Bulletin (Department of Ichthyology, Rhodes University, Grahamstown), 
no. 32, pp. 635-682, 13 figs., pis. 103-104. 

S0RENSEN, WlLLLVM 

1884. Om lydorganer hos fiske. En physiologisk og comparativ-anatomisk unders0- 
gelse. University of Copenhagen, pp. 1-245, 4 pis. 

1897. Some remarks on Dr. Thilo's memoire on "Die Umbildungen an den Gliedmassen 

der Fische." Morphologisches Jahrbuch (Leipzig), no. 25, pp. 170-189, 6 figs. 
Thllo, Otto 

1896. Die Umbildungen an den Gliedmassen der Fische. Morphologisches Jahrbuch, 
no. 24, pp. 287-355, 7 figs., pis. 6-9. 

1898. Erganzungen zu meiner Abhandlung "Die Umbildungen an den Gliedmassen der 

Fische." Morphologisches Jahrbuch, no. 26, pp. 81-90. 



PROCEEDINGS 

OF THE 

CALIFORNIA ACADEMY OF SCIENCES 

FOURTH SERIES 



Festschrift for George Sprague Myers 



Vol. XXXVIII, No. 22, pp. 411-414; 1 fig. December 31, 1970 



AMPUTATION AND REPLACEMENT OF 
MARGINAL SPINES IN CTENOID PERCOID 

SCALES 

By 

Howard McCully 
Divison of Systematic Biology, Stanford University, Stanford California 94305 

In the ctenoid scales of most primitive percoid fishes the spines are found only 
on the free edge of the scale where it is not covered by another scale. Typically 
each spine with its base is a separate bone (a scalelet) fixed to the fiber layer of 
the scale. The fiber layer forms a flexible joint between adjacent scalelets. Cycloid 
scales in some genera, for example Ryptkus and Grammistes, have homologous 
scalelets that lack spines; in other genera, for example Siniperca, the posterior 
fields lack scalelets that are homologous with the spines of ctenoid scales. 

Posterior growth in the scales that have spines is by increments of single 
scalelets. Except for rare and inconsistent specimens or species, new scalelets do 
not form radially to another unless at least the tip of the spine has been lost or 
amputated. When two spines outgrow, the one between them (the shorter and 
older spine) tends to lie flatter than it did and the tip is amputated by osteoclasis. 
Nearly always, solution pits can be seen on the end of the stump. Then a new 
scalelet will be laid down distal to the stump. Eventually the new one will grow 
until it extends beyond its neighbors and they in turn will be amputated and 
replaced. 

The fully grown scalelets with their spines stand erect or nearly so and hold 

[411] 



412 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

up the connecting fiber layer and the overlying soft tissues of the epidermis and 
scale pocket. This forms a ridge usually in the arc of a circle that projects from 
the side of the fish. Except at the free edge, the posterior fields of such scales are 
covered with the amputated stumps of scalelets that once bore the spines that 
made the scale ctenoid. 

The figure shows all the steps in the life cycle of one of these marginal spines 
except for some intermediate sizes in the growth of the scalelet. It shows part of 
the free margin of the posterior field of a scale, 2.79 mm. long, from the flank of 
a pike-perch, Stizostedion canadense (Smith), SU-S673. It is unusual in that it 
shows the amputated tip of a scalelet still in position. Usually such a tip is lost 
soon after it is cut off. Most of the scales from this fish show 1 or 2 scalelets in 
this condition. It is the only fish I have yet encountered that showed any tips of 
amputated spines and the beginning of the new scalelets beneath them. 

The scalelet that has just been amputated is the marginal one that does not 
reach as far back as the others. The original length of this scalelet was 0.165 mm. 
The tip is 0.059 mm. long. The gap from which the bone was removed is 0.013 
mm. wide. The primary ossification of the forming scalelet is wider than the 
old tip and lies beneath it. The new ossification measures 0.027 mm. in the 
anteroposterior diameter and is 0.045 mm. wide. Eventually the new scalelet will 
extend beyond the two beside it and they in their turn will be amputated and 
replaced. 

The specimen was cleaned of as much adhering soft tissue as possible, stained 
with Alizarin Red S., and mounted in air under a cover slip secured with a few 
drops of polyvinl chloride glue. The mount is so made that it dries under pres- 
sure and the free edge of the scale is held down. The gaps between the scalelets 
shrink on drying and are now narrower than they were in life. In spite of the 
cleaning, a layer or two of cells lies over the bony tissues in most of the figure. 

Williamson (1851) was the first to notice that the posterior field of perch, 
Perca, scales were made up of the bases of spines that were broken off. Baudelot 
(1873) saw that the perfect spines were only at the margins and that all spines 
not marginal were broken (brisee). He concluded, as had Williamson, that the 
spines were formed at the free edge. Neither one is able to explain how the scale- 
lets became broken and Bauelot says that more observations are needed. Hase 
(1911) studied perch, young of the year, and reached the mistaken conclusion 
that the posterior scalelets were formed near the nucleus and pushed out toward 
the margin. At the margin they then grew their spines. There was no clear ex- 
planation of what happened until I completed my doctoral research (McCully, 
1961). 

From my examination of this and material from the Serranidae I conclude: 

1. That the small size of the amputation gap means that only a few cells 
can be excreting the osteoclastic material. 



Vol. XXXVIII] McCULLY: MARGINAL SPINES IN CTENOID SCALES 



413 




Figure 1. Posterior margin of scale from the flank of a pike-perch, Stizostedion cana- 
dense. (See text for explanation.) 



2. Nearby cells must be protecting the bone that is not attacked. 

3. The material removed from the bone may be redeposited nearby. 

4. There is a regulating mechanism that can differentiate the excretory 
activities of a few selected cells from their neighbors. Another regulatory 



414 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

change stops the unusual secretion and returns the cells, presumably, to their 
former state. 

5. It is possible that migratory cells are the source of osteoclastic excretion. 

6. There is, in a very small area, exposed to external observation and 
manipulation the whole of the sequence of bone growth and absorption. This 
type of scale offers a convenient way to test the action of drugs in an intact 
animal on any aspect of the physiological processes of bone growth and ab- 
sorption except for those peculiar to the replacement of cartilage. 

REFERENCES 

Baudelot, Emile 

1873. Recherches sur la structure et la developpment des ecailles des poissons osseux. 
Archives de Zoologie Experimentale et Generale, vol. 2, pp. 87-244 and pp. 429- 
480, pis. v-xi. 
Hase, Albrecht 

1911. Die morphologische Entwicklung der Ctenoidschuppe. Anatomischer Anzeiger, 
1911, vol. 40, pp. 337-356, 28 figs. 
McCuLLY, Howard 

1961. The comparative anatomy of the scales of the Serranid fishes. University Micro- 
films, Inc., Ann Arbor, Michigan, 248 pp. 
Williamson, William Crawford 

1851. Investigations into the structure and development of the scales and bones of fishes. 
Philosophical Transactions of the Royal Society of London, 1851, pp. 643-702. 



PROCEEDINGS 

OF THE 

CALIFORNIA ACADEMY OF SCIENCES 

FOURTH SERIES 



Festschrift for George Sprague Myers 



Vol. XXXVIII, No. 23, pp. 415-420; 3 figs.; 1 table. December 31, 1970 



SIZE AND DISTRIBUTION OF PROTEINS 
IN ELASMOBRANCH PLASMA 

By 

Robert J. Heckly 

Naval Biomedical Research Laboratory, Naval Sitpl)ly Center 
Oakland, California 94625 

and 

Earl S. Herald 

Steinhart Aciiiariiim, California Academy of Sciences 
Golden Gate Park, San Francisco, Ccdifornia 9411S 

As part of a project sponsored by the Office of Naval Research (X0014-67- 
C-0343 ) some members of the staff at Steinhart Aquarium embarked on an elec- 
trophoretic investigation of the classification value of elasmobranch plasma 
proteins. Samples from 11 of the 27 species available were sent to the Naval 
Biomedical Research Laboratory for ultracentrifuge analysis. The results show 
species and perhaps family differences not suspected from cellulose acetate and 
acrylimide gel electrophoresis. 

Similarities in blood constituents have been used by many investigators to 
show relationship between various animals. Genetic changes and natural selec- 
tion are believed to affect serum or plasma constituents less than the gross anat- 
omy or other such features. Thus, similarities in composition of blood or serum 
would indicate a relationship; certainly if they differed significantly it would 
indicate that the individuals are not closely related. Such classification of elasmo- 
branchs has been the subject of immunologic or electrophoretic analyses (Clem 

[415] 



416 



CALIFORNIA ACADEMY OF SCIENCES 



[Proc. 4th Ser. 




Figure 1. Schlieren patterns obtained from plasma sample of 696 cm. adult male basking 
shark (Cetorhinus maximus). Photographs were taken at indicated times after rotor attained 
full speed (59,780 rpm). Numbers below major peaks in last two frames refer to observed 
sedimentation coefficients. 



and Small, 1967; Irisawa and Irisawa, 1954; Rasmussen and Rasmussen, 1967; 
Shuster and Goodman, 1968) but no comparative study of molecular size of 
serum proteins has been made. This report describes the results of some ultra- 
centrifugal analyses of blood plasma from several selected species. 

MATERIALS AND METHODS 

Blood was usually obtained by caudal puncture using Sequester-Sol (a 
dipotassiumethylenediaminetetracetate supplied by Cambridge Chemical Prod- 
ucts, Inc., Detroit) and for most samples, immediately centrifuged; the plasma 
was then removed and stored at 2° C. Analyses were made with a Spinco Model 
E Analytical ultracentrifuge fitted with schlieren optics using 12 mm. cells. The 
plasma were analyzed at the highest practical concentrations, either 1:2 or 1:4, 



F 


49.5 


3.1 


3.4 


8 


13 


— 


F 


64.7 


1.9 


3 


7 


— 


17 


M 


75.7 


5.3 


4 


7 


— 


17 


M 


696.0 


— 


4 


7 


— 


18 


F 


94.0 


3.4 


4.5 


7 


— 


17 


F 


119.3 


2.75 


4.3 


7 


14 


17 


F 


— 


— 


4 


7 


— 


17 


F 


89.0 


5.9 


5.6 


8 


12 


17 


F 


57.2 


0.64 


4 


8 


— 


17 


M 


78.6 


— 


5 


S 


12 


17 


M 


76.2 


— 


5 


11 


14 


— 


M 


90.2 


— 


5 


11 


14 


— 



Vol. XXXVIII] HECKLY & HERALD: ELASMOBRANCH PLASMA 417 

Table 1. Sedimentation coefficients of the principal plasma proteins. 

CM Total Total 
Common Name Scientific Name Sex Length Protein Component S Values 

Pacific lamprey, Lampetra tridentata 
Sevengill, Xotorynchus maculatus 
Horn shark, Heterodontus francisci 
Basking shark, Cetorhinns maximus 
Swell shark, Cephaloscy Ilium ventriosum 
Leopard shark, Triakis semifasciata 
Dogfish, Squalus acanthias suckleyi 
Shovelnose guitarfish, Rhinobatos product us 
Thornback ray, Platyrhinoidis triseriata 
Pacific electric ray, Torpedo californica 
Big skate, Raja binoculata (A) 
Big skate, Raja binoculata (B) 

SO that minor constituents would be detected. In all instances the centrifuge 
was operated at 59,780 rpm. 

RESULTS 

Figure 1 shows a series of photographs taken during the sedimentation of 
serum from a basking shark. The first three frames clearly show the separation 
and sedimentation of the 18 S component. However, even after 48 minutes the 
4 S component had not been completely resolved from the 7 S proteins. Since the 
area under the curve is proportional to the concentration of that component it is 
clear that the major protein in this serum had a high molecular weight. If the 7 S 
material is a globular protein, its molecular weight is probably in excess of 
160,000. 

The results of ultracentrifuge analyses of plasma from a number of animals 
are summarized in figures 2 and 3 and the observed sedimentation coefficients 
are listed in table 1. The relative concentrations of the various components 
in each plasma can be estimated from the schlieren patterns. 

Except for the sevengill shark and thornback ray, the relative concentration 
of macroglobulin (17 S component) in shark plasma is higher than in mam- 
malian serum. As is evident in figure 2, the 7 S globulins in human serum have 
not been resolved from the albumin, but of course, continued centrifugation 
did separate the 7 S proteins from the albumin. Similarly, low molecular weight 
proteins were resolved from the 6 to 8 S components in the other plasma on pro- 
longed centrifugation indicating that there is at least a small amount of 
albumin-sized material in all of the elasmobranch plasma, even though this is not 
evident in the 32-minute frames shown in figures 2 and 3. 

In addition to indicating relative concentration, the schlieren patterns pro- 
vide a measure of homogeneity of the protein components in that if all mol- 



418 



CALIFORNIA ACADEMY OF SCIENCES 



[Proc. 4th Ser. 




Figure 2. Schlieren patterns obtained with various plasma 32 minutes after rotor at- 
tained full speed, 59,780 rpm. Numbers given below patterns indicate approximate sedimen- 
tation coefficients of components represented by peaks. 



Vol. XXXVIII] HECKLY & HERALD: ELASMOBRANCH PLASMA 



419 



^ SHOVELNOSE 
GUITAR FISH 




Figure 3. Schlieren patterns obtained with various plasma il minutes after rotor at- 
tained full speed, 59,780 rpm. Numbers given below patterns indicate approximate sedimen- 
tation coefficients of components represented by peaks. 



420 CALIFORNIA ACADEMY OF SCIENCES [Proc. 4th Ser. 

ecules are the same size the schHeren peak will be sharp, such as is demonstrated 
by the patterns for the skate and ray. A broad peak indicates either heteroge- 
neity of the component with respect to sedimentation velocity or a high diffusion 
coefficient. For instance, albumin, even though it is homogenous, would result 
in a rather broad peak after 32 minutes. 

The lamprey is unusual in that it appears to have a higher relative concen- 
tration of albumin (the 3.4 S component) than the elasmobranchs, and in this 
respect it is more like mammalian serum because, as is well known, albumin 
is the principal constituent in normal mammalian serum. In contrast, as is 
shown in figure 1, the principal protein in the basking shark was of the 7 S 
variety. 

Skates appear to lack 17 S protein in their plasma but they have high con- 
centrations of both the 10 and 14 S proteins. Although the patterns from dif- 
ferent skates do not yield identical patterns the variations that have been ob- 
served are shown in the last two photographs of figure 2. Since the presence or 
absence of a component is perhaps more significant than small differences in 
concentration, it is of interest that neither the lamprey nor the big skate have 
17 S proteins and that they both have proteins in the 8 to 14 S range. 

Also of interest are the differences in the component S values between the 
guitarfish and the thornback. These two species were formerly placed in sep- 
arate families now combined in a single family (Rhinobatidae). It would seem 
logical that all family members would have similar sedimentation coefficients, 
yet in this case the shovelnose has a 12 S component which is strangely lacking 
in the thornback. Further study will be needed to evaluate this difference. 

As is shown in figures 2 and 3 and in table 1, the leopard shark, shovelnose, 
guitarfish, and electric ray have four distinct S components in their plasma, 
whereas all the other eight species studied have only three major components. 
With the exception of the guitarfish and thornback all species are represent- 
atives of individual families not closely related. 

LITERATURE CITED 

Clem, L. W., and P. A. Small 

1967. Phylogeny of immunoglobulin structure and function. I. Immunoglobulins 
of the lemon shark. Journal of Experimental Medicine, vol. 125, pp. 893-920. 
Irisawa, H., and a. F. Irisawa 

1954. Blood serum protein of the marine Elasmobranchii. Science, vol. 120, pp. 849-851. 
Rasmussen, L. E., and R. A. Rasmussen 

1967. Comparative protein and enzyme profiles of the cerebrospinal fluid, extradural 

fluid, nervous tissue and sera of elasmobranchs. Chapt. 25, pp. 361-379. 
In Perry W. Gilbert (Ed.) Sharks, skates and rays. Baltimore, Johns Hop- 
kins Press, 624 pp. 
Shuster, J., and J. W. Goodman 

1968. Phylogenetic studies of shark immunoglobulins. Nature, vol. 219, pp. 298-299. 



INDEX TO VOLUME XXXVIII 



FOURTH SERIES 



New names and principal reference in boldface type 



Ablepharus, 190 
bivittatus, 190 

bivittatus lindbergi, 170, 190 
grayanus, 169, 190 
pannonicus, 170, 190, 191 
Acanthodactylus, 186 

cantoris, 170, 186 
Acanthogobius flavimanus, 207, 208, 209, 210, 

211, 213, 237 
Acanthogobius flavimanus in the San Fran- 
cisco Bay-Delta region of California, 
Explosive spread of the oriental goby, 
by Martin R. Brittan, John D. Hop- 
kirk, Jerrold D. Conners and Michael 
Martin, 207-214 
{Acanthuridae) , The dorsal and anal spine- 
locking apparatus of surgeon fishes, 
by James C. Tyler, 391-410 
Acanthurus, 392, 393, 403 
monroviae, 397 

triostegus, 392, 393, 394, 395, 396, 397, 
400, 402, 403, 406, 407, 408 
Acerina, 255 
Acestridium, 158 

discus, 157, 158, 159 
Acestridium discus Haseman, near Manaus, 
Brazil, Rediscovery of the loricariid 
catfish, by Robert L. Hasstir, 157-162 
Adinia, 292, 295 
xenica, 295 
Afghanistan, a checklist and key to the 
herpetofauna, The amphibians and 
reptiles of, by Alan E. Leviton and 
Steven C. Anderson, 163-206 
Agama, 176 

agilis, 167, 176 
agrorensis, 168, 176 
badakhshana, 168, 176 
caucasica, 168, 176 
erythrogastra, 168, 177 
himalayana, 168, 177 
himalayana himalayana, 177 



lehmanni, 168, 177 
megalonyx, 178 
nupta, 168, 177 
nuristanica, 168, 178 
ruderata, 178 
ruderata baluchiana, 178 
ruderata megalonyx, 167, 178 
scutellata, 181 
tuberculata, 168, 178 
versicolor, 178 
Agamura, 182 

femoralis, 166, 182, 205 
persica, 166, 182 
Agkistrodon, 201 
halys, 172,201 
himalayanus, 172, 201 
mokasen, 201 
Ahaetulla prasina. 111, 114 
Ailanthus, 2 

Alcala, Angel C, see Brown, Walter C. 
Alosa sapidissima, 212 
Alsophylax pipiens, 166, 182, 205 
American characoid fishes, with special ref- 
erence to Probolodus heterostomus, 
Scale-eating, by Tyson R. Roberts, 
383-390 
Amiva gabbiana, 284 

amphibians and reptiles of Afghanistan, a 
checklist and key to the herpetofauna, 
The, by Alan E. Leviton and Steven 
C. Anderson, 163-206 
Amputation and replacement of marginal 
spines in ctenoid percoid scales, by 
Howard McCully, 411-414 
a)ial spine-locking apparatus of surgeon 
fishes, (Acanthuridae), The dorsal 
and, by James C. Tyler, 391-410 
Ancistrodon, 201, 202 
halys, 201 
himalayanus, 201 
Anderson, Steven C ., see Leviton, Alan E. 
Anguilla, 256 



[421] 



422 



CALIFORNIA ACADEMY OF SCIENCES 



[Proc. 4th Ser. 



Anguis 

lumbricalis, 201 

ventralis, 181 
Annotated chronological bibliography of the 
publications of George Sprague Myers 
(to the end of 1969), 19-52 
Anolis 

capito, 284 

copei, 284 

humilis, 279 

intermedius, 279 

oxylophus, 284 

pachypus, 284 

trochilus, 284 
Ansonia 

mcgregori, 109 

muelleri, 109, 356 
Aphredoderus, 215, 226, 229, 232, 234, 235, 
258, 259, 296 

sayanus, 217, 229, 231, 235, 236, 239, 
294, 295 
Aphyocharax, 388 
Aplopeltura boa, 114 
Apogon, 215, 241, 244, 254, 255 

astradorsatus, 217, 245 

mclanotaenia, 255 

apogonid fishes, Some nerve patterns and 

their systematic significance in parac- 

anthopterygian, salmoniform, gobioid, 

and, by Warren C. Freihofer, 215-264 

Aprionodon, 82 

isodon, 67 
Archolaemus, 267, 269, 270, 271 

blax, 266, 267, 268, 270, 271 
Argyropelecus, 234, 237 
Argyropleura, 150 
Astronotus ocellatus, 389 
Astyanax, 388, 389 

fasciatus, 388 
Atelopus varius, 283 
Aulopus, 256 
Avocettina, 99 



Barbourula, 117 

busuangensis, 109 

Basciliscus 

mitratus, 279 
plumifrons, 284 
vittatus, 284 



Bassozetus, 236 
Bathygobius, 237 

lineatus, 217, 237, 249, 254 
Bathylagus alascanus, 249 
Batrachuperus, 174 

mustersi, 165, 174 
Batrachyperus mustersi, 174 
Belonophago, 384 

hutsebouti, 384 
Bembrops, 377 

bibliography of the publications of George 
Sprague Myers {to the end of 1969), 
Annotated chronological, 19-52 
Blcpharosteres grayanus, 190 
Boa 

imperator, 284 

johnii, 194 

tatarica, 195 

turcica, 194 
Bohlke, James E., A new species of the 
doradid catfish genus Leptodoras, with 
comments on related forms, 53-62 
Boiga, 195 

angulata, 110, 115 

cynodon, 115 

dendrophila, 115 

drapiezi, 115 

philippina, 115 

trigonata melanocephalus, 173, 195 
Boreogadus, 227 

saida, 217 
Bothriechis nigroviridis, 285 
Bothriopsis proboscideus, 285 
Bothrops atrox, 285 
Brachymeles, 116, 117, 121, 125, 128 

bonitae, 112, 116, 117 

cebuensis, 116, 117 

elerae, 112, 116 

gracilis, 112, 116, 117 

hilong, 113, 116 

pathfinderi, 112, 116 

samarensis, 113, 116, 117 

schadenbergi, 113, 116, 117 

talinis, 113, 116 

tridactylus, 110, 113, 116, 117 

vermis, 116, 117 

wrighti, 113, 116 
Brazil, A new gymnotoid fish from the Rio 
Tocantins, by Maarten Korringa, 265- 
272 



Vol. XXXVIII] 



INDEX 



423 



Brazil, Rediscovery of the loricariid catfish, 

Acestridium discus Haseman, near 

Manaus, by Robert L. Hassiir, 157- 

162 

Briggs, John C, Tropical shelf zoogeography, 

131-138 
Briltan, Martin R., John D. Hopkirk, Jer- 
rold D. Conners, and Michael Martin, 
Explosive spread of the oriental goby 
Acanthogobius flavimanus in the San 
Francisco Bay-Delta region of Cali- 
fornia, 207-214 
Brotula, 252, 256, 258 

clarkae, 217, 222, 223, 240 
multibarbata, 222 
Brotuloides emmalas, 217 
Brown, Walter C, and Angel C. Alcala, The 
zoogeography of the herpetofauna of 
the Philippine Islands, a fringing 
archipelago, 105-130 
Bufo, 174 

agua, 283 

andersonii, 165, 174 
auritus, 283 
biporcatus, 109 
chilensis, 279 
gabbi, 283 

haematiticus, 279, 283 
margaritifer, 278 
melanochloris, 283 
pleuropterus, 278 
simus, 278 
typhonius, 279 
valliceps, 283 
veraguensis, 278, 279 
viridis, 165, 174 
vulgaris, 174, 279 
Bungarus caeruleus, 206 
Bunopus, 182 

tuberculatus, 166, 182 



Calamaria, 111 

bitorques, 114 

gervaisi, 114 

lumbricoidea, 114 

palawanensis, 114 

virgulata, 114 
California, Explosive spread of the oriental 
goby Acanthogobius flavimanus in the 



San Francisco Bay-Delta region of, by 
Martin R. Brittan, John D. Hopkirk, 
Jerrold D. Conners, and Michael 
Martin, 207-214 
Callionymus, 368, 369, 370, 371, 372, 373, 
374, 375,376,377 

decoratus, 367, 376 

flagris, 366, 367, 368, 369, 372, 376 
Calliophis calligaster, 115 
Calliscyllium, 87, 89, 90, 93 

venustum, 87, 89, 90 
Calotes, 178 

cristatellus, 112 

marmoratus, 112 

versicolor, 163, 167, 178 
Carcharhinus, 67, 69, 80, 82 

acronotus, 67 

albimarginatus, 67 

altimus, 67 

amblyrhynchus, 67 

borneensis, 67 

cauta, 67 

falciformis, 67 

galapagensis, 67 

leucas, 67 

limbatus, 67 

longimanus, 67 

maculipinnis, 67 

melanopterus, 67 

menisorrah, 67 

milberti, 67 

obscurus, 67 

pleurotaenia, 67 

porosus, 67 

remotus, 67 

sorrah, 67 

springeri, 67 

tjutjot, 67 

velox, 67 
Catastoma psephotum, 285 
catfish, Acestridium discus Haseman, near 
Manaus, Brazil, Rediscovery of the 
loricariid, by Robert L. Hassur, 157- 
162 
Catoprion, 383, 386, 387, 388 
Catostomus occidentalis, 212 
Caularchus, 364, 374 

maeandricus, 364 
Central America, On the trail of the golden 



424 



CALIFORNIA ACADEMY OF SCIENCES 



[Proc. 4th Ser. 



frog: with Warszewicz and Gabb in, 
by Jay M. Savage, 273-288 
Centrotrachelus asmussi, 181 
Cephaloscyllium ventriosum, 417 
Cepola rubescens, 255 
Ceramodactylus affinus, 186 
Cerastes persicus, 202 
Cetorhinus maximus, 416, 417 
Chaenogaleus, 82 
Chalcidolepis metallicus, 284 
Chaperina fusca, 110 

characid fish, Hysteronotus myersi, from 
Peru, A new species of glandulocau- 
dine, by Stanley H. Weitzman and 
Jamie E. Thomerson, 139-156 
characoid fishes, with special reference to 
Probolodus heterostomus, Scale-eating 
American, by Tyson R. Roberts, 383- 
390 
Charax, 388 

Chauliodus macouni, 237 
Cheimarrichthys, 376, 378, 379 
Cheirodon, 388 
Chelopus 

funerus, 285 

gabbii, 285 
Chologaster, 258 

papilliferus, 217, 229 
Chrossomus neogaeus, 2 
Chrysopelea paradisi, 111, 114 
Cinosternum leucostomum, 285 
Cleome, 2 

Clevelandia ios, 211 
Clupea pallasii, 250 
Coelorhynchus scaphopsis, 217 
Cohen, Daniel M., How many recent fishes 

are there? , 341-346 
Coluber, 164, 195 

aulicus, 197 

bitorquatus, 198 

blumenbachii, 199 

collaris, 197 

constrictor, 195 

dione, 197 

halys, 201 

irregularis, 195 

karelinii, 173, 195 

lebetinus, 203 

mucosus, 199 

ravergieri, 173, 196 



rhodorhachis, 173, 196 
schokari, 199 
sibilans, 199 

(Taphrometopon) lineolatus, 199 

ventromaculatus, 173, 196 
C ompagno , L. J. V., Systeniatics of the genus 
Hemitriakis (Selachii: Carcharhini- 
dae) , and related genera, 63-98 
Coniophanes fissidens, 284 
Conners, Jerrold D., see Brittan, Martin R. 
Contia 

bicolor, 197 

pachyura, 285 
Coregonus, 250 
Cornufer, 104 
Coronella 

taeniolata, 198 

tessellata, 198 
Corynopoma, ISO 
Corythophanes cristatus, 284 
Cosymbotus platyurus, 111, 112 
Cranopsis fastidiosus, 283 
Crenichthys, 292, 295, 296 

baileyi, 293 
Crepidius epioticus, 283 
Crossobamon, 182 

eversmanni, 166, 182 

lumsdeni, 166, 183 

maynardi, 166, 183 
Ctenochaetus, 392 

ctenoid percoid scales, Amputation and re- 
placement of marginal spines in, by 
Howard McCully, 411-414 
Ctenolucius, 383 
Cursoria elegans, 194 
Cyclocorus, 117, 118, 119, 128 

lineatus, 114, 118 

nuchalis, 359 
Cyclophis 

aestivus, 279 

persicus, 197 
Cyprinodon, 292, 293 

diabolis, 15 

variegatus, 293 
Cyrtodactylus, 164, 166, 183, 185 

agusanensis, 112 

annulatus, 112 

caspius, 167, 183, 184 

fedtschenkoi, 167, 184, 185 

macularius, 185 



Vol. XXXVIII] 



INDEX 



425 



philippinicus, 112 
pulchellus, 1S3 
redimiculus, 112 
russowii, 167, 184 
scaber, 167, 184 
watsoni, 167, 184 



Dasia 

griff ini, 113 
olivaceum, 113 
smaragdina, 111, 113 
Dendrelaphis 

caudolineatus, 114 
pictus. 111, 114 
Dendrobates 

lugubris, 278 
pumilio, 278 
speciosus, 278 
talamancae, 283 
tinctorius, 283 
typographus, 283 
Dendrophidium melanotropis, 284 
Description of a new subspecies of Rhab- 
dophis auriculata in the Philippines, 
with comments on the zoogeography 
of Mindanao Island, by Alan E. Levi- 
ton, 347-362 
DeWitt, Hugh H., A revision of the fishes of 
the genus Notothenia from the New 
Zealand region, including Macquarie 
Island, 299-340 
Dibamus argenteus, 112 
Dicrolene, 237, 252, 256 

intronigra, 217, 234, 235, 246 
kanazawi, 217 
Dinematichthys, 258 

iluocoeteoides, 217, 222 
Diplogrammus goramensis, 367 
Dirrhizodon, 82 

doradid catfish genus Leptodoras, with com- 
ments on related forms, A new species 
of, by James E. Bohlke, 53-62 
Dormitator, 259 
Dorosoma petenensis, 212 
Dorsal and anal spine-locking apparatus of 
surgeon fishes (Acanthuridae) , The, 
by James C. Tyler, 391-410 
Draco 

bimaculatus, 112 



everetti, 112 

mindanensis, 112 

ornatus, 112 

quadrisi, 112 

rizali, 112 

volans. 111, 112 
Draconetta, 365, 367, 368, 369, 370, 371, 372, 
373,374, 375,376,377 

acanthopoma, 366, 367, 368, 369, 372, 
373, 374, 375 

oregona, 373 
Draculo, 378 

Drymobius boddaertii, 284 
Dryocalamus 

subannulatus, 114 

tristrigatus, 114 
Dryophiops philippina, 114 



Earle, Sylvia A., see Mead, Giles W. 
Echis, 202 

carinatus, 172,202 
Eigenmannia, 269, 270, 271 

conirostris, 271 

macrops, 271 

virescens, 270 
Eirenis, 197 

persica, 172, 197 
Elaphe, 197 

bairdii, 15 

dione, 173, 197 

erythrura. 111, 114 

parreysii, 197 
Elaps circinalis, 285 

elasmobranch plasma, Size and distribution 
of proteins in, by Robert J. Heckly 
and Earl S. Herald, 415-420 
Elassoma zonatum, 295 
Eleginus, 227 

gracilus, 217 
Eleotris, 253 

fuscus, 217, 253, 254 
Emoia 

atrocostata. 111, 113 

caeruleocauda, 113 

ruficauda, 113 
Empetrichthys, 295, 296 
Enneacanthus, 2 
Epigeichthys, 251 
Epigonus robustus, 244 



426 



CALIFORNIA ACADEMY OF SCIENCES 



[Proc. 4th Ser. 



Eremias, 187 

acutirostris, 171, 187 

aporosceles, 169, 170, 187 

aria, 171, 187 

fasciata, 170, 187 

grammica, 171, 187 

guttulata, 188 

guttulata watsonana, 171, 188 

intermedia, 171, 188 

lineolata, 170, 188 

(Mesalina) watsonana, 188 

nigrocellata, 171, 188 

persica, 189 

regeli, 170, 189 

(Rhabderemias) scripta, 189 

(Scapteira) acutirostris, 187 

(Scapteira) aporosceles, 187 

(Scapteira) grammica, 187 

scripta, 170, 189 

velox, 189 

velox persica, 171, 189 

velox velox, 171, 189 
Eridacnis, 87, 88, 89, 90, 91, 92, 93 

alcocki, 90, 93 

barbouri, 67, 72,90, 93 

radcliffei, 67, 90, 93 

sinuans, 65, 67, 90, 93 
Eristicophis, 202 

macmahoni, 172, 202 
Erythrolamprus venustissimus, 284 
Eryx, 164, 171, 194 

elegans, 171, 194 

jaculus, 171, 195 

jaculus czarewskii, 194 

johnii, 172, 194, 195 

miliaris, 171, 172, 194 

persicus, 195 

tataricus, 172, 195 
Etheostoma, 294, 295, 296 

asprigene, 294 

radiosum, 294, 295 

spectabile, 294, 295 
Eublepharis, 185 

hardwiciiii, 185 

macularius, 165, 185 
Eucynopotamus, 388 
Eugaleus, 75 
Eugnathichthys, 384 
Eumeces, 191 

b